Tunnel lining concrete thickness detection method, system, equipment and storage medium

Abstract

The present application discloses a method, system, device and storage medium for detecting the thickness of tunnel lining concrete, obtaining the first structural type, first geological condition information and first lining material information of the tunnel to be detected; detecting the tunnel to be detected by ground penetrating radar equipment to obtain first radar data; inputting the first structural type, the first geological condition information, the first lining material information and the first radar data into a trained concrete thickness detection model, detecting the lining concrete thickness of the tunnel to be detected by the concrete thickness detection model, and obtaining a first thickness detection result; wherein the concrete thickness detection model is built based on the UNeXt network architecture. This method can improve the detection accuracy of the thickness of tunnel lining concrete, and supports automated implementation, with high detection efficiency, which can help reduce the cost of detection. The present application can be widely used in the field of artificial intelligence technology.

Description

Technical Field

[0001] The present application relates to the field of artificial intelligence technology, and in particular to a method, system, device and storage medium for detecting the thickness of tunnel lining concrete. Background Art

[0002] Tunnel lining concrete refers to a layer of concrete covering the surface of the inner wall of the tunnel, which is used to strengthen the tunnel structure and protect the inner wall. This layer of concrete usually has a certain thickness to provide sufficient support and protection. The thickness information of tunnel lining concrete is very important, which can evaluate the safety and stability of the tunnel structure. If the thickness of the concrete is insufficient, it may lead to insufficient structural strength and increase the risk of structural damage. Moreover, by detecting the thickness of the lining concrete, it can provide important data support for tunnel maintenance and repair work, which helps to optimize engineering design and maintenance plans.

[0003] In related technologies, when detecting the thickness of tunnel lining concrete, mathematical modeling is generally performed on the tunnel lining concrete, and the obtained simulation data is studied, such as obtaining thickness information through relevant deep learning algorithms. However, in actual applications, it is found that the environment of tunnel construction is relatively complex, and existing algorithms often have the problem of low accuracy. In addition, modeling and other applications have certain requirements on the experience and skills of operators, and the cost is relatively high.

[0004] Therefore, the problems existing in the prior art still need to be solved and optimized. Summary of the invention

[0005] The purpose of this application is to solve one of the technical problems existing in the related art to at least a certain extent.

[0006] To this end, an object of embodiments of the present application is to provide a method, system, device and storage medium for detecting the thickness of tunnel lining concrete.

[0007] In order to achieve the above technical objectives, the technical solutions adopted by the embodiments of the present application include:

[0008] On the one hand, an embodiment of the present application provides a method for detecting the thickness of a tunnel lining concrete, the method comprising:

[0009] Acquire a first structural type, first geological condition information, and first lining material information of a tunnel to be inspected;

[0010] Detecting the tunnel to be detected by using a ground penetrating radar device to obtain first radar data;

[0011] Inputting the first structure type, the first geological condition information, the first lining material information and the first radar data into a trained concrete thickness detection model, detecting the lining concrete thickness of the tunnel to be detected by using the concrete thickness detection model, and obtaining a first thickness detection result;

[0012] Among them, the concrete thickness detection model is built based on the UNeXt network architecture.

[0013] In addition, a tunnel lining concrete thickness detection system according to the above embodiment of the present application may also have the following additional technical features:

[0014] Further, in one embodiment of the present application, the concrete thickness detection model is trained by the following steps:

[0015] Acquire a second structural type, second geological condition information, second lining material information and label information of a plurality of training tunnels; wherein the label information is used to mark the actual result of the thickness of the lining concrete in the training tunnel;

[0016] Detecting the training tunnel by using a ground penetrating radar device to obtain second radar data;

[0017] inputting the second structure type, the second geological condition information, the second lining material information and the second radar data into a concrete thickness detection model to be optimized, detecting the lining concrete thickness of the training tunnel by using the concrete thickness detection model, and obtaining a second thickness detection result;

[0018] Determining a training loss value according to the second thickness detection result and the label information;

[0019] According to the loss value, the parameters of the concrete thickness detection model are updated to obtain a trained concrete thickness detection model.

[0020] Further, in one embodiment of the present application, after the step of obtaining the second radar data, the method further includes a preprocessing process;

[0021] The preprocessing process includes: performing at least one of data cleaning processing, data normalization processing, image enhancement processing, cropping processing, data expansion processing, and category balancing processing on the second radar data.

[0022] Furthermore, in one embodiment of the present application, the image enhancement processing includes at least one of brightness adjustment, contrast adjustment, rotation, and flipping operations.

[0023] Further, in one embodiment of the present application, determining the loss value of training according to the second thickness detection result and the label information includes:

[0024] A training loss value is determined by Dice Loss or Focal Loss according to the second thickness detection result and the label information.

[0025] Furthermore, in one embodiment of the present application, detecting the tunnel to be detected by a ground penetrating radar device to obtain first radar data includes:

[0026] Determining a setting position and a scanning direction of the ground penetrating radar device according to the first structure type;

[0027] Determining acquisition parameters of the ground penetrating radar device according to the first geological condition information and the first lining material information;

[0028] According to the setting position, the scanning direction and the acquisition parameters, the ground penetrating radar device is controlled to detect the tunnel to be detected to obtain first radar data.

[0029] Furthermore, in one embodiment of the present application, the acquisition parameters include radar frequency, pulse width, and sampling interval.

[0030] On the other hand, an embodiment of the present application provides a tunnel lining concrete thickness detection system, the system comprising:

[0031] An acquisition unit, used to acquire a first structural type, first geological condition information and first lining material information of a tunnel to be detected;

[0032] A detection unit, configured to detect the tunnel to be detected by using a ground penetrating radar device to obtain first radar data;

[0033] a processing unit, configured to input the first structure type, the first geological condition information, the first lining material information, and the first radar data into a trained concrete thickness detection model, detect the lining concrete thickness of the tunnel to be detected by using the concrete thickness detection model, and obtain a first thickness detection result;

[0034] Among them, the concrete thickness detection model is built based on the UNeXt network architecture.

[0035] On the other hand, an embodiment of the present application provides a computer device, including:

[0036] at least one processor;

[0037] at least one memory for storing at least one program;

[0038] When the at least one program is executed by the at least one processor, the at least one processor implements the above-mentioned method for detecting the thickness of tunnel lining concrete.

[0039] On the other hand, an embodiment of the present application further provides a computer-readable storage medium, which stores a program executable by a processor. When the program executable by the processor is executed by the processor, it is used to implement the above-mentioned method for detecting the thickness of a tunnel lining concrete.

[0040] The advantages and benefits of the present application will be partially given in the following description, and partially become apparent from the following description, or be understood through the practice of the present application:

[0041] The embodiment of the present application discloses a method for detecting the thickness of tunnel lining concrete, which obtains the first structural type, first geological condition information and first lining material information of the tunnel to be detected; detects the tunnel to be detected by ground penetrating radar equipment to obtain first radar data; inputs the first structural type, the first geological condition information, the first lining material information and the first radar data into a trained concrete thickness detection model, detects the lining concrete thickness of the tunnel to be detected by the concrete thickness detection model, and obtains a first thickness detection result; wherein the concrete thickness detection model is built based on the UNeXt network architecture. This method can improve the detection accuracy of the thickness of tunnel lining concrete, and supports automated implementation, with high detection efficiency, which can help reduce the cost of detection. BRIEF DESCRIPTION OF THE DRAWINGS

[0042] In order to more clearly illustrate the embodiments of the present application or the technical solutions in the prior art, the following introduction is made to the drawings of the embodiments of the present application or the related technical solutions in the prior art. It should be understood that the drawings introduced below are only for the convenience of clearly describing some embodiments of the technical solutions of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without paying any creative work.

[0043] Figure 1 A schematic diagram of an implementation environment of a tunnel lining concrete thickness detection method provided in an embodiment of the present application;

[0044] Figure 2 A schematic diagram of a process for detecting the thickness of a tunnel lining concrete provided in an embodiment of the present application;

[0045] Figure 3 A schematic diagram of the structure of a tunnel lining concrete thickness detection system provided in an embodiment of the present application;

[0046] Figure 4A schematic diagram of the structure of a computer device provided in an embodiment of the present application. DETAILED DESCRIPTION

[0047] In order to make the purpose, technical solutions and advantages of the present application clearer, the present application is further described in detail below in conjunction with the accompanying drawings and examples. It should be understood that the specific embodiments described herein are only used to explain the present application and are not intended to limit the present application. When the following description refers to the drawings, unless otherwise indicated, the same numbers in different drawings represent the same or similar elements. The embodiments described in the following exemplary embodiments do not represent all embodiments consistent with the embodiments of the present application. They are only examples of devices and methods consistent with some aspects of the embodiments of the present application as detailed in the attached claims.

[0048] It is understood that the terms "first", "second", etc. used in this application can be used to describe various concepts in this article, but unless otherwise specified, these concepts are not limited by these terms. These terms are only used to distinguish one concept from another concept. For example, without departing from the scope of the embodiment of the present application, the first information may also be referred to as the second information, and similarly, the second information may also be referred to as the first information. Depending on the context, the words "if" and "if" as used herein can be interpreted as "at the time of" or "when" or "in response to determination".

[0049] The terms "at least one", "multiple", "each", "any", etc. used in this application, at least one includes one, two or more, multiple includes two or more, each refers to each of the corresponding multiple, and any refers to any one of the multiple.

[0050] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as those commonly understood by those skilled in the art to which this application belongs. The terms used herein are only for the purpose of describing the embodiments of this application and are not intended to limit this application.

[0051] First, some nouns involved in this application are analyzed:

[0052] Ground-Penetrating Radar (GPR) is a non-destructive testing technology that detects underground structures and features by sending electromagnetic waves into the ground and recording the echoes. A ground-penetrating radar system usually consists of a transmitter and a receiver, which scans the underground through a mobile or fixed measurement position on the ground. When electromagnetic waves encounter material interfaces with different dielectric constants, part of the energy is reflected back, and these reflected signals are captured by the receiver and used to generate images and data of underground structures. Ground-penetrating radar can be used for a variety of applications, including geological exploration, soil analysis, building inspection, archaeological research, pipeline inspection, etc. It can provide information about the depth, shape, size and material properties of underground objects, helping users to conduct underground exploration and analysis. Ground-penetrating radar is widely used in engineering, scientific research and other fields, providing users with a non-invasive, fast and accurate method of underground detection.

[0053] Artificial intelligence (AI) is a new technical science that studies and develops theories, methods, technologies and application systems for simulating, extending and expanding human intelligence. AI is a branch of computer science that attempts to understand the essence of intelligence and produce a new type of intelligent machine that can respond in a similar way to human intelligence. Research in this field includes robotics, speech recognition, image recognition, natural language processing and expert systems. AI can simulate the information process of human consciousness and thinking. AI is also a theory, method, technology and application system that uses digital computers or machines controlled by digital computers to simulate, extend and expand human intelligence, perceive the environment, acquire knowledge and use knowledge to obtain the best results.

[0054] Machine Learning (ML): It is a multi-disciplinary subject involving probability theory, statistics, approximation theory, convex analysis, algorithm complexity theory and other disciplines. It specializes in studying how computers simulate or implement human learning behavior to acquire new knowledge or skills and reorganize existing knowledge structures to continuously improve their performance. Machine learning is the core of artificial intelligence and the fundamental way to make computers intelligent. Its applications are spread across all areas of artificial intelligence. Machine learning usually includes artificial neural networks, belief networks, reinforcement learning, transfer learning, inductive learning, and learning by teaching.

[0055] Deep Learning (DL) is a new research direction in the field of machine learning. It is introduced into machine learning to make it closer to the original goal - artificial intelligence. Deep learning is to learn the inherent laws and representation levels of sample data. The information obtained in the learning process is of great help in the interpretation of data such as text, images and sounds. Its ultimate goal is to enable machines to have analytical learning capabilities like humans and to recognize data such as text, images and sounds.

[0056] Tunnel lining concrete refers to a layer of concrete covering the surface of the inner wall of the tunnel, which is used to strengthen the tunnel structure and protect the inner wall. This layer of concrete usually has a certain thickness to provide sufficient support and protection. The thickness information of tunnel lining concrete is very important, which can evaluate the safety and stability of the tunnel structure. If the thickness of the concrete is insufficient, it may lead to insufficient structural strength and increase the risk of structural damage. Moreover, by detecting the thickness of the lining concrete, it can provide important data support for tunnel maintenance and repair work, which helps to optimize engineering design and maintenance plans.

[0057] In related technologies, when detecting the thickness of tunnel lining concrete, mathematical modeling is generally performed on the tunnel lining concrete, and the obtained simulation data is studied, such as obtaining thickness information through relevant deep learning algorithms. However, in actual applications, it is found that the environment of tunnel construction is relatively complex, and existing algorithms often have the problem of low accuracy. In addition, modeling and other applications have certain requirements on the experience and skills of operators, and the cost is relatively high.

[0058] In view of this, a method for detecting the thickness of a tunnel lining concrete is provided in an embodiment of the present application. The method can improve the detection accuracy of the thickness of a tunnel lining concrete, support automation, have high detection efficiency, and can help reduce the cost of detection.

[0059] Next, the implementation environment of a tunnel lining concrete thickness detection method provided in an embodiment of the present application is first introduced. Figure 1 , Figure 1A schematic diagram of an implementation environment of a method for detecting the thickness of a tunnel lining concrete is provided. The software and hardware main bodies of the implementation environment mainly include a terminal device 110 and a server 120, and the terminal device 110 is in communication connection with the server 120. The method for detecting the thickness of a tunnel lining concrete can be configured on the terminal device 110 side or on the server 120 side. For example, when the method for detecting the thickness of a tunnel lining concrete is configured on the terminal device 110 side, an application for detecting the thickness of a tunnel lining concrete can be realized by relying solely on the terminal device 110; when the method for detecting the thickness of a tunnel lining concrete is configured on the server 120 side, an application for detecting the thickness of a tunnel lining concrete can be realized through the interaction between the terminal device 110 and the server 120.

[0060] Specifically, the terminal device 110 in the present application may include but is not limited to any one or more of a smart watch, a smart phone, a computer, a personal digital assistant (PDA), an intelligent voice interaction device, a smart home appliance or a vehicle-mounted terminal. The server 120 may be an independent physical server, or a server cluster or distributed system composed of multiple physical servers, or a cloud server that provides basic cloud computing services such as cloud services, cloud databases, cloud computing, cloud functions, cloud storage, network services, cloud communications, middleware services, domain name services, security services, CDN (Content Delivery Network), and big data and artificial intelligence platforms. The terminal device 110 and the server 120 may establish a communication connection through a wireless network or a wired network, which uses standard communication technology and / or protocols. The network may be set to the Internet or any other network, such as but not limited to a local area network (LAN), a metropolitan area network (MAN), a wide area network (WAN), a mobile, wired or wireless network, a private network or a virtual private network.

[0061] Of course, it is understandable that Figure 1 The implementation environment in the embodiment of the present application is only an optional application scenario of a tunnel lining concrete thickness detection method provided in the embodiment of the present application, and the actual application is not fixed to Figure 1 The hardware and software environment shown below. Figure 1 The implementation environment shown is used to describe in detail a method for detecting the thickness of a tunnel lining concrete provided in an embodiment of the present application.

[0062] Please refer to Figure 2 , Figure 2is a flow chart of a method for detecting the thickness of a tunnel lining concrete provided in an embodiment of the present application, referring to Figure 2 The present application provides a method for detecting the thickness of tunnel lining concrete, including but not limited to:

[0063] Step 210: obtaining a first structural type, first geological condition information, and first lining material information of the tunnel to be inspected;

[0064] Step 220: Detect the tunnel to be detected by ground penetrating radar equipment to obtain first radar data;

[0065] Step 230: input the first structure type, the first geological condition information, the first lining material information and the first radar data into a trained concrete thickness detection model, detect the lining concrete thickness of the tunnel to be detected by the concrete thickness detection model, and obtain a first thickness detection result;

[0066] Among them, the concrete thickness detection model is built based on the UNeXt network architecture.

[0067] In an embodiment of the present application, a method for detecting the thickness of a tunnel lining concrete is provided. When detecting the thickness of a tunnel lining concrete, the tunnel to be detected is recorded as a tunnel to be detected. In an embodiment of the present application, there is no restriction on the location and size of the tunnel to be detected, which can be selected according to actual needs. Then, the structural type, geological condition information, and lining material information of the tunnel to be detected can be obtained, which are recorded as the first structural type, the first geological condition information, and the first lining material information, respectively.

[0068] In the embodiment of the present application, when the thickness of the lining concrete in the tunnel to be detected is detected, the corresponding first structure type, first geological condition information and first lining material information are obtained. These information can provide diversified and comprehensive parameters, which facilitates the subsequent concrete thickness detection model to detect the thickness result more accurately. Among the above information, the structure type is used to characterize the type of the tunnel structure, for example, it can include a circular tunnel, a rectangular tunnel, a horseshoe tunnel or a multi-section combined tunnel. When the tunnel to be detected belongs to a multi-section combined tunnel, the specific type of each section can be recorded and then spliced ​​as the structure type corresponding to the multi-section combined tunnel. The geological condition information is used to characterize the geological conditions of the tunnel to be detected, such as hard rock geology, soft soil geology, aquifer or fault geology, etc. Under different geological conditions, the detection of radar signals is different. For example, under hard rock geology, the reflection signal characteristics of the lining concrete and the rock interface are distinct; under soft soil geology, the geological characteristics of the soil may cause radar signal attenuation or scattering; aquifers or fault geology belong to special geological structures, and they have a greater impact on radar signals. The lining material information is used to characterize the material category of the lining concrete used in the tunnel to be inspected, such as ordinary concrete, special concrete or composite concrete, etc. This application does not limit the specific categories contained in the above information.

[0069] It can be understood that in the embodiments of the present application, introducing the structure type, geological condition information, and lining material information into the algorithm can improve the algorithm's recognition ability in complex geological backgrounds.

[0070] In the embodiment of the present application, the tunnel to be detected can also be detected by a ground penetrating radar device, and the obtained radar data is recorded as the first radar data. Then, the first structure type, the first geological condition information, the first lining material information and the first radar data are input into the trained concrete thickness detection model, and the lining concrete thickness of the tunnel to be detected is detected by the concrete thickness detection model, and the obtained thickness detection result is recorded as the first thickness detection result.

[0071] In the embodiment of the present application, for the concrete thickness detection model, it can select the lightweight convolutional neural network architecture UNeXt, which has fewer parameters and computational complexity and is suitable for resource-constrained scenarios; secondly, the use of deep separable convolutional layers instead of traditional convolutional layers can reduce parameters and computational complexity while maintaining the accuracy of the model; the introduction of multi-layer perceptrons (MLP) combined with convolution operations can extract richer local-global features. In the decoder part, the UNeXt network uses a deep separable deconvolution layer to achieve effective feature upsampling while reducing model parameters, and adds jump connections to help the model capture feature information of different scales and improve the perception of subtle structures. The model is pruned and quantized to further reduce parameters and computational complexity while maintaining performance. Through these designs, a lightweight model can be constructed that is suitable for the tunnel lining concrete thickness detection task of ground penetrating radar, achieving efficient reasoning in a resource-constrained environment, while having low computational costs and high performance.

[0072] In some embodiments, the concrete thickness detection model is trained by the following steps:

[0073] Acquire a second structural type, second geological condition information, second lining material information and label information of a plurality of training tunnels; wherein the label information is used to mark the actual result of the thickness of the lining concrete in the training tunnel;

[0074] Detecting the training tunnel by using a ground penetrating radar device to obtain second radar data;

[0075] inputting the second structure type, the second geological condition information, the second lining material information and the second radar data into a concrete thickness detection model to be optimized, detecting the lining concrete thickness of the training tunnel by using the concrete thickness detection model, and obtaining a second thickness detection result;

[0076] Determining a training loss value according to the second thickness detection result and the label information;

[0077] According to the loss value, the parameters of the concrete thickness detection model are updated to obtain a trained concrete thickness detection model.

[0078] In the embodiment of the present application, before using the concrete thickness detection model, it is necessary to train it to improve its prediction performance. Specifically, when training the concrete thickness detection model, batch training data can be obtained. In the embodiment of the present application, multiple other tunnels can be determined and recorded as training tunnels, and then the second structure type, second geological condition information, second lining material information and label information of the training tunnel are obtained, wherein the content of the second structure type, second geological condition information, and second lining material information is similar to the aforementioned embodiment and will not be repeated here. Here, the label information can be manually marked to record the information of the real result of the thickness of the lining concrete in the training tunnel. In the embodiment of the present application, the second radar data corresponding to the training tunnel is obtained by detection through ground penetrating radar equipment, and the second structure type, second geological condition information, second lining material information and second radar data are input into the concrete thickness detection model to be optimized, and the predicted result is recorded as the second thickness detection result.

[0079] After obtaining the second thickness detection result, the prediction accuracy of the concrete thickness detection model can be evaluated according to the label information and the second thickness detection result, so as to facilitate iterative updating of the parameters of the concrete thickness detection model. Specifically, a loss value can be determined according to the label information and the second thickness detection result, and according to the loss value, the concrete thickness detection model can be back-propagated and trained to update the relevant parameters inside it.

[0080] Specifically, for various models in the field of artificial intelligence, the accuracy of its prediction can be measured by the loss function. The loss function is defined on a single training data and is used to measure the prediction error of a training data. Specifically, the loss value of the training data is determined by the label of the single training data and the prediction result of the model for the training data. In actual training, a training data set includes a lot of training data, so the cost function is generally used to measure the overall error of the training data set. The cost function is defined on the entire training data set and is used to calculate the average value of the prediction error of all training data, which can better measure the prediction effect of the model. For general machine learning models, based on the aforementioned cost function, plus the regularization term that measures the complexity of the model, it can be used as the objective function of the training. Based on this objective function, the loss value of the entire training data set can be calculated. There are many types of commonly used loss functions, such as 0-1 loss function, square loss function, absolute loss function, logarithmic loss function, cross entropy loss function, etc., which can all be used as loss functions of machine learning models, and will not be elaborated one by one here. Exemplarily, in an embodiment of the present application, Dice Loss or Focal Loss can be used. Dice Loss is a commonly used loss function used to measure the similarity between the model prediction results and the true labels. It is commonly used in image segmentation tasks, and is particularly suitable for processing datasets with imbalanced categories. Dice Loss calculates the ratio between the intersection and union of the prediction results and the true labels, and optimizes the model by minimizing this ratio. Focal Loss is a loss function used to process imbalanced categories and difficult-to-classify samples. It introduces an adjustable parameter γ to adjust the weights of the loss function for easy-to-classify samples and difficult-to-classify samples. FocalLoss can pay more attention to these samples when processing difficult-to-classify samples, thereby improving the performance of the model. These two loss functions are lightweight loss functions, which are conducive to achieving lightweight models.

[0081] In some embodiments, after the step of obtaining the second radar data, the method further includes a preprocessing process;

[0082] The preprocessing process includes: performing at least one of data cleaning processing, data normalization processing, image enhancement processing, cropping processing, data expansion processing, and category balancing processing on the second radar data.

[0083] In the embodiment of the present application, after obtaining the second radar data, it can be preprocessed. Preprocessing may include performing data cleaning, data normalization, image enhancement, cropping, data expansion, and category balancing on the second radar data, wherein data cleaning is to ensure the integrity and accuracy of the data by checking the abnormal samples, missing values, or wrong annotations in the data set and performing corresponding processing. Standardization or normalization of the data is to make the data have similar scales and ranges, which helps to speed up the convergence of the model and improve the generalization ability of the model to different features. Image enhancement is also an important preprocessing step, which can increase the diversity of data and the robustness of the model by adjusting brightness, contrast, rotation, flipping, etc. Subsequently, according to the input requirements of the model, the image is cropped and resized to ensure that the model can process fixed-size input images. In addition, it is also common to increase the number of samples in the data set using data expansion techniques, such as random cropping, scaling, translation, etc., to increase the diversity of data and the generalization ability of the model. In addition, in order to avoid the model's preference for certain categories, it is necessary to ensure that the number of samples in each category in the dataset is balanced through category balancing, which can be achieved through methods such as oversampling, undersampling, or synthesizing minority class samples.

[0084] In some embodiments, detecting the tunnel to be detected by using a ground penetrating radar device to obtain first radar data includes:

[0085] Determining a setting position and a scanning direction of the ground penetrating radar device according to the first structure type;

[0086] Determining acquisition parameters of the ground penetrating radar device according to the first geological condition information and the first lining material information;

[0087] According to the setting position, the scanning direction and the acquisition parameters, the ground penetrating radar device is controlled to detect the tunnel to be detected to obtain first radar data.

[0088] In an embodiment of the present application, when the tunnel to be detected is detected by a ground-penetrating radar device, the setting position and scanning direction of the ground-penetrating radar device can be determined according to the first structure type, so that the required radar data can be easily detected. In addition, the acquisition parameters of the ground-penetrating radar device can be determined according to the first geological condition information and the first lining material information, which may specifically include radar frequency, pulse width, sampling interval, etc., and the present application does not limit this. Then, after determining the setting position, scanning direction and acquisition parameters, the detection task can be performed to collect the first radar data.

[0089] It can be understood that in the embodiments of the present application, the structure type, geological condition information and lining material information can be used in a composite manner, based on which the radar data collection plan is determined to improve the radar data collection accuracy and coverage effect, and they are input into the model together with the collected radar data, which can facilitate the model to achieve more accurate detection based on these mutually related and influential data, thereby improving data utilization efficiency and detection accuracy.

[0090] Below, in combination with a specific application process, the implementation of a tunnel lining concrete thickness detection method provided in an embodiment of the present application is introduced and illustrated.

[0091] (1) Determine the diversified collection area:

[0092] In the embodiment of the present application, the tunnel area where data collection is required is determined, and it is ensured that the area contains various types of concrete structures and linings of different thicknesses to reflect the actual engineering situation.

[0093] Consider including areas that may have defects or damage, which helps the model learn to handle the diversity of lining structures. Diversity types include:

[0094] Diversity of geological conditions: Consider tunnel structures under different geological conditions, including different types of strata, rocks, soils, etc. The differences in geological conditions may affect the propagation and reflection characteristics of the GPR signal, so it is necessary to cover various geological conditions in the dataset.

[0095] Variation of concrete thickness: The thickness of tunnel lining concrete may vary depending on factors such as tunnel location, purpose, etc. Therefore, the dataset should contain samples of concrete linings of different thicknesses to ensure that the model can make accurate predictions for data within a different range of thicknesses.

[0096] Diversity of lining materials: In addition to concrete, tunnel linings may include other materials such as steel bars, polymers, etc. Considering that different materials may have different GPR reflection characteristics, it is necessary to include samples with different material compositions in the dataset.

[0097] Diversity of tunnel morphology: Tunnels may have different morphologies, such as straight tunnels, curved tunnels, slope tunnels, etc. Different morphologies of tunnel structures will affect the propagation path and reflection characteristics of the GPR signal, so it is necessary to cover tunnels of various morphologies in the dataset.

[0098] (2) Determine the acquisition parameters:

[0099] According to the requirements of the task and the actual geological conditions, determine the appropriate acquisition parameters, such as radar frequency, pulse width, sampling interval, etc. Considering the impact of different concrete thicknesses on radar signals, adjust the parameters to obtain clear lining structure information.

[0100] (3) Collect data:

[0101] Use the ground penetrating radar system to collect data in the determined area. Ensure complete coverage of the tunnel lining area that needs to be segmented, and ensure that the data covers different structures and thickness changes. Adjust the position and scanning direction of the radar according to the actual situation to obtain comprehensive lining information.

[0102] (4) Record ground conditions:

[0103] While collecting data, record the actual ground conditions, including the type of lining, thickness, possible defects or damage, etc. These records will serve as a reference for subsequent data processing and model verification, and help verify the model's adaptability to real conditions.

[0104] (5) Ground Penetrating Radar Data Preprocessing:

[0105] Data cleaning: Check the dataset for abnormal samples, missing values, or incorrect annotations and handle them accordingly. This may involve deleting or repairing corrupted samples to ensure the integrity and accuracy of the data.

[0106] Data standardization / normalization: Standardize or normalize the data to make them have similar scale and range. This helps to speed up the convergence of the model and improve the generalization ability of the model to different features.

[0107] Image enhancement: Enhance the images in the dataset, such as adjusting brightness, contrast, rotation, flipping, etc., to increase the diversity of the data and the robustness of the model.

[0108] Cropping and resizing: Crop and resize the images according to the model’s input requirements. Typically, neural network models require fixed-size input images, so all images need to be resized to the same size.

[0109] Data augmentation: Use data augmentation techniques to increase the number of samples in a data set, such as random cropping, scaling, translation, and other operations, to increase data diversity and the generalization ability of the model.

[0110] Category balancing: Ensure that the number of samples in each category in the data set is balanced to avoid the model's preference for certain categories. This can be achieved through methods such as oversampling, undersampling, or synthesizing minority class samples.

[0111] Data labeling: Based on the information recorded on the ground, the data is labeled at the pixel level, and each pixel is labeled as the corresponding concrete thickness category to form a training data set.

[0112] Dataset partitioning: Divide the dataset into training set, validation set and test set. The ratio usually adopted is 70% of the data for training, 15% for validation, and 15% for testing to evaluate the performance and generalization ability of the model.

[0113] (6) Build a lightweight tunnel lining concrete thickness detection model

[0114] Lightweight Convolutional Neural Network Architecture: Select a lightweight convolutional neural network (CNN) architecture. Here, UNeXt is selected as the tunnel lining concrete thickness detection model.

[0115] UNeXt is a lightweight medical image segmentation network, but it not only performs well in medical images, but also has good segmentation performance in other images with boundaries. UNeXt is an image segmentation network based on convolutional multilayer perceptron (MLP) with an encoder-decoder architecture. It combines the characteristics of convolution and MLP, reduces the computational workload and storage requirements of the model, and aims to improve the performance and efficiency of image segmentation, making it suitable for resource-constrained environments.

[0116] There are several reasons for choosing UNeXt network for tunnel lining concrete thickness detection:

[0117] (6.1) Lightweight design: The UNeXt network adopts a lightweight design, which reduces the amount of computation and storage requirements while maintaining high performance. This is very beneficial for real-time or fast analysis at the tunnel site, which can reduce hardware requirements and increase processing speed.

[0118] (6.2) Efficient segmentation performance: The UNeXt network has achieved good performance in the field of medical image segmentation, indicating that it is also highly applicable to tasks such as tunnel lining concrete thickness detection. The network can accurately identify and segment the thickness information of the concrete lining, providing reliable data support for engineers.

[0119] (6.3) Easy to improve and optimize: The UNeXt network has good scalability and flexibility, and can be easily integrated with other sensor data and prior knowledge to further improve the accuracy and reliability of segmentation. In addition, with the continuous development of deep learning technology, the UNeXt network can also be continuously optimized and updated to adapt to new application requirements and data changes.

[0120] In the UNeXt network, the input image is first processed by the encoder. The first 3 blocks of the encoder are convolution blocks, which are used to extract local features of the image. The next 2 blocks are tokenized MLP blocks, which are used to capture global features. This combination of convolution and MLP can capture both local and global features of the image. This comprehensive feature extraction method enables the network to more accurately identify the thickness information of the tunnel lining concrete and improve the accuracy of segmentation. The decoder part contains 2 tokenized MLP blocks and 3 convolution blocks. Each decoder block uses bilinear interpolation to increase the feature resolution to restore the segmentation result of the same size as the original image. The role of the decoder is to maximize the restoration of pixel-level segmentation results based on the features extracted by the encoder. In addition, the UNeXt network also contains jump connections between the encoder and the decoder. These connections help to combine low-level features of the tunnel lining concrete thickness with high-level features, further improving the accuracy and performance of concrete thickness detection. In addition, the UNeXt network can reduce the amount of calculation and parameters by using tokenized MLP blocks and special channel axis displacement operations, while improving the robustness and generalization of the network, which means that the network can cope with the challenges of different complex geological conditions, tunnel structure types and lining materials, and has good generalization performance. Therefore, the use of the UNeXt network for tunnel lining concrete thickness detection has the advantages of high efficiency, accuracy, flexibility and adaptability, and provides a reliable and efficient solution for the task of concrete thickness detection in tunnel engineering.

[0121] Reference Figure 3 , a tunnel lining concrete thickness detection system proposed in an embodiment of the present application includes:

[0122] An acquisition unit 310 is used to acquire a first structural type, first geological condition information and first lining material information of a tunnel to be detected;

[0123] The detection unit 320 is used to detect the tunnel to be detected by using a ground penetrating radar device to obtain first radar data;

[0124] The processing unit 330 is used to input the first structure type, the first geological condition information, the first lining material information and the first radar data into a trained concrete thickness detection model, detect the lining concrete thickness of the tunnel to be detected by using the concrete thickness detection model, and obtain a first thickness detection result;

[0125] Among them, the concrete thickness detection model is built based on the UNeXt network architecture.

[0126] It can be understood that the contents of the above method embodiments are all applicable to the present system embodiments, the functions specifically implemented by the present system embodiments are the same as those of the above method embodiments, and the beneficial effects achieved are also the same as those achieved by the above method embodiments.

[0127] Reference Figure 4 , an embodiment of the present application provides a computer device, including:

[0128] at least one processor 410;

[0129] At least one memory 420, used to store at least one program;

[0130] When at least one program is executed by at least one processor 410, at least one processor 410 implements Figure 2 A method for detecting the thickness of tunnel lining concrete is shown.

[0131] Similarly, the contents of the above method embodiments are all applicable to the present computer device embodiment. The functions specifically implemented by the present computer device embodiment are the same as those of the above method embodiments, and the beneficial effects achieved are also the same as those achieved by the above method embodiments.

[0132] The embodiment of the present application also provides a computer-readable storage medium, which stores a program executable by the processor 410. When the program executable by the processor 410 is executed by the processor 410, it is used to execute the above-mentioned method for detecting the thickness of tunnel lining concrete.

[0133] The present application also discloses a computer-readable storage medium in which a program executable by a processor is stored. When the program executable by the processor is executed by the processor, it is used to implement the following Figure 2 An embodiment of a method for detecting thickness of tunnel lining concrete is shown.

[0134] It is understandable that if Figure 2 The contents of the embodiment of a tunnel lining concrete thickness detection method shown in the figure are applicable to the embodiment of this computer-readable storage medium. The functions specifically implemented by this computer-readable storage medium embodiment are similar to those in the embodiment of the present invention. Figure 2 The tunnel lining concrete thickness detection method shown in the embodiment is the same as that shown in the embodiment, and the beneficial effects achieved are the same as those of the embodiment shown in the embodiment. Figure 2 The beneficial effects achieved by the embodiment of the tunnel lining concrete thickness detection method shown are also the same.

[0135] In some selectable embodiments, the function / operation mentioned in the block diagram may not occur in the order mentioned in the operation diagram. For example, depending on the function / operation involved, the two boxes shown in succession can actually be executed substantially simultaneously or the boxes can sometimes be executed in reverse order. In addition, the embodiment presented and described in the flow chart of the application is provided by way of example, for the purpose of providing a more comprehensive understanding of technology. The disclosed method is not limited to the operation and logic flow presented herein. Selectable embodiments are expected, wherein the order of various operations is changed and the sub-operation of a part described as a larger operation is performed independently.

[0136] In addition, although the present application is described in the context of functional modules, it should be understood that, unless otherwise specified, one or more of the functions and / or features can be integrated into a single physical system and / or software module, or one or more functions and / or features can be implemented in a separate physical system or software module. It can also be understood that a detailed discussion of the actual implementation of each module is unnecessary for understanding the present application. More specifically, in view of the properties, functions and internal relationships of the various functional modules in the system disclosed herein, the actual implementation of the module will be understood within the conventional techniques of engineers. Therefore, those skilled in the art can implement the present application set forth in the claims without excessive experimentation using ordinary techniques. It can also be understood that the specific concepts disclosed are merely illustrative and are not intended to limit the scope of the present application, which is determined by the full scope of the attached claims and their equivalents.

[0137] If the function is implemented in the form of a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present application, or the part that contributes to the prior art or the part of the technical solution, can be embodied in the form of a software product, which is stored in a storage medium and includes several instructions for a computer device (which can be a personal computer, a server, or a network device, etc.) to perform all or part of the steps of the various embodiments of the present application. The aforementioned storage medium includes: U disk, mobile hard disk, read-only memory (ROM, Read-Only Memory), random access memory (RAM, Random Access Memory), disk or optical disk, and other media that can store program codes.

[0138] The logic and / or steps represented in the flowchart or otherwise described herein, for example, can be considered as an ordered list of executable instructions for implementing logical functions, and can be embodied in any computer-readable medium for use by an instruction execution system, system or device (such as a computer-based system, a system including a processor, or other system that can fetch instructions from an instruction execution system, system or device and execute the instructions), or in conjunction with these instruction execution systems, systems or devices. For the purpose of this specification, "computer-readable medium" can be any system that can contain, store, communicate, propagate or transport a program for use by an instruction execution system, system or device, or in conjunction with these instruction execution systems, systems or devices.

[0139] More specific examples of computer-readable media (a non-exhaustive list) include the following: an electrical connection with one or more wires (electronic systems), a portable computer disk case (magnetic systems), a random access memory (RAM), a read-only memory (ROM), an erasable and programmable read-only memory (EPROM or flash memory), a fiber optic system, and a portable compact disk read-only memory (CDROM). In addition, the computer-readable medium may even be a paper or other suitable medium on which the program is printed, since the program may be obtained electronically, for example, by optically scanning the paper or other medium, followed by editing, deciphering or, if necessary, processing in another suitable manner, and then stored in a computer memory.

[0140] It should be understood that the various parts of the present application can be implemented by hardware, software, firmware or a combination thereof. In the above-mentioned embodiments, multiple steps or methods can be implemented by software or firmware stored in a memory and executed by a suitable instruction execution system. For example, if implemented by hardware, as in another embodiment, it can be implemented by any one of the following technologies known in the art or their combination: a discrete logic circuit having a logic gate circuit for implementing a logic function for a data signal, a dedicated integrated circuit having a suitable combination of logic gate circuits, a programmable gate array (PGA), a field programmable gate array (FPGA), etc.

[0141] In the above description of this specification, the description with reference to the terms "one embodiment / example", "another embodiment / example" or "certain embodiments / examples" etc. means that the specific features, structures, materials or characteristics described in conjunction with the embodiment or example are included in at least one embodiment or example of the present application. In this specification, the schematic representation of the above terms does not necessarily refer to the same embodiment or example. Moreover, the specific features, structures, materials or characteristics described may be combined in any one or more embodiments or examples in a suitable manner.

[0142] Although the embodiments of the present application have been shown and described, those skilled in the art will appreciate that various changes, modifications, substitutions and variations may be made to the embodiments without departing from the principles and spirit of the present application, and that the scope of the present application is defined by the claims and their equivalents.

[0143] The above is a specific description of the preferred implementation of the present application, but the present application is not limited to the embodiments. Those skilled in the art can make various equivalent modifications or substitutions without violating the spirit of the present application. These equivalent modifications or substitutions are all included in the scope defined by the claims of the present application.

[0144] In the description of this specification, the description with reference to the terms "one embodiment", "another embodiment" or "certain embodiments" etc. means that the specific features, structures, materials or characteristics described in conjunction with the embodiments or examples are included in at least one embodiment or example of the present application. In this specification, the schematic representation of the above terms does not necessarily refer to the same embodiment or example. Moreover, the specific features, structures, materials or characteristics described may be combined in any one or more embodiments or examples in a suitable manner.

[0145] Although the embodiments of the present application have been shown and described, those skilled in the art will appreciate that various changes, modifications, substitutions and variations may be made to the embodiments without departing from the principles and spirit of the present application, and that the scope of the present application is defined by the claims and their equivalents.

Claims

1. A method for detecting the thickness of tunnel lining concrete, characterized in that: The method comprises: Acquire a first structural type, first geological condition information, and first lining material information of a tunnel to be inspected; Detecting the tunnel to be detected by using a ground penetrating radar device to obtain first radar data; Inputting the first structure type, the first geological condition information, the first lining material information and the first radar data into a trained concrete thickness detection model, detecting the lining concrete thickness of the tunnel to be detected by using the concrete thickness detection model, and obtaining a first thickness detection result; Wherein, the concrete thickness detection model is built based on the UNeXt network architecture; The detecting the tunnel to be detected by using a ground penetrating radar device to obtain first radar data includes: Determining a setting position and a scanning direction of the ground penetrating radar device according to the first structure type; Determining acquisition parameters of the ground penetrating radar device according to the first geological condition information and the first lining material information; According to the setting position, the scanning direction and the acquisition parameters, the ground penetrating radar device is controlled to detect the tunnel to be detected to obtain first radar data.

2. A method for detecting thickness of tunnel lining concrete according to claim 1, characterized in that: The concrete thickness detection model is trained by the following steps: Acquire a second structural type, second geological condition information, second lining material information and label information of a plurality of training tunnels; wherein the label information is used to mark the actual result of the thickness of the lining concrete in the training tunnel; Detecting the training tunnel by using a ground penetrating radar device to obtain second radar data; inputting the second structure type, the second geological condition information, the second lining material information and the second radar data into a concrete thickness detection model to be optimized, detecting the lining concrete thickness of the training tunnel by using the concrete thickness detection model, and obtaining a second thickness detection result; Determining a training loss value according to the second thickness detection result and the label information; According to the loss value, the parameters of the concrete thickness detection model are updated to obtain a trained concrete thickness detection model.

3. A method for detecting thickness of tunnel lining concrete according to claim 2, characterized in that: After the step of obtaining the second radar data, the method further includes a preprocessing process; The preprocessing process includes: performing at least one of data cleaning processing, data normalization processing, image enhancement processing, cropping processing, data expansion processing, and category balancing processing on the second radar data.

4. A method for detecting thickness of tunnel lining concrete according to claim 3, characterized in that: The image enhancement process includes at least one of brightness adjustment, contrast adjustment, rotation, and flipping operations.

5. A method for detecting thickness of tunnel lining concrete according to claim 2, characterized in that: Determining a loss value of training according to the second thickness detection result and the label information includes: A training loss value is determined by Dice Loss or Focal Loss according to the second thickness detection result and the label information.

6. A method for detecting thickness of tunnel lining concrete according to claim 1, characterized in that: The acquisition parameters include radar frequency, pulse width, and sampling interval.

7. A tunnel lining concrete thickness detection system, characterized in that: The system comprises: An acquisition unit, used to acquire a first structural type, first geological condition information and first lining material information of a tunnel to be detected; A detection unit, configured to detect the tunnel to be detected by using a ground penetrating radar device to obtain first radar data; a processing unit, configured to input the first structure type, the first geological condition information, the first lining material information, and the first radar data into a trained concrete thickness detection model, detect the lining concrete thickness of the tunnel to be detected by using the concrete thickness detection model, and obtain a first thickness detection result; Wherein, the concrete thickness detection model is built based on the UNeXt network architecture; The detecting the tunnel to be detected by using a ground penetrating radar device to obtain first radar data includes: Determining a setting position and a scanning direction of the ground penetrating radar device according to the first structure type; Determining acquisition parameters of the ground penetrating radar device according to the first geological condition information and the first lining material information; According to the setting position, the scanning direction and the acquisition parameters, the ground penetrating radar device is controlled to detect the tunnel to be detected to obtain first radar data.

8. A computer device, characterized in that: include: at least one processor; at least one memory for storing at least one program; When the at least one program is executed by the at least one processor, the at least one processor implements a tunnel lining concrete thickness detection method as described in any one of claims 1-6.

9. A computer-readable storage medium storing a program executable by a processor, characterized in that: The processor executable program is used to implement a tunnel lining concrete thickness detection method as described in any one of claims 1 to 6 when executed by the processor.

Images