Urban traffic infrastructure operation state detection and evaluation method

Abstract

The invention discloses an urban traffic infrastructure operation state detection and evaluation method, and belongs to the technical field of traffic infrastructure monitoring. In order to achieve rapid and intelligent identification and evaluation of urban traffic infrastructure operation state indexes, the method comprises the steps that the traffic infrastructure operation state indexes are set, and then a data set used for urban traffic infrastructure operation state detection and evaluation is constructed; constructing an intelligent identification model of the traffic infrastructure operation state indexes, wherein the intelligent identification model comprises a basic model, a first auxiliary model, a second auxiliary model and a strengthening model; using the obtained data set for urban traffic infrastructure operation state detection and evaluation to train the constructed basic model, the first auxiliary model, the second auxiliary model and the strengthening model, and using the obtained trained basic model as a final traffic infrastructure operation state index intelligent identification model to construct a digital twin model; and a traffic infrastructure operation state evaluation method is constructed, and the traffic infrastructure operation state is evaluated.

Description

Technical Field

[0001] The present invention belongs to the technical field of traffic infrastructure monitoring, and in particular relates to a method for detecting and evaluating the operating status of urban traffic infrastructure. Background Art

[0002] Urban transportation infrastructure includes roads, bridges, tunnels and other facilities. The safety of transportation infrastructure is directly related to the safety of public life and property. It can help identify early diseases and damage, take necessary maintenance and reinforcement measures, and extend the service life of transportation facilities. Timely maintenance and repair can prevent small problems from turning into major failures, avoid unnecessary high maintenance costs and facility replacement costs. Reduce traffic congestion and delays caused by poor facility conditions, improve traffic efficiency, and ensure transportation safety.

[0003] At present, the inspection of transportation infrastructure is carried out independently according to the type of facility, that is, road inspection, bridge inspection, and tunnel inspection are carried out separately, resulting in independent data, making it difficult to consider the mutual influence and linkage between different facilities, leading to the neglect of potential problems. Data and inspection results of different types of facilities are difficult to share effectively, resulting in information islands and affecting the overall operation and maintenance efficiency. At the same time, the inspection scope and content between different facilities may be repeated, leading to repeated inspections and waste of resources, increasing manpower and material costs. Summary of the invention

[0004] The problem to be solved by the present invention is to quickly and intelligently identify and evaluate the operating status indicators of urban transportation infrastructure, and proposes a method for detecting and evaluating the operating status of urban transportation infrastructure.

[0005] To achieve the above object, the present invention is implemented through the following technical solutions: A method for detecting and evaluating the operating status of urban transportation infrastructure comprises the following steps: S1. Using a vehicle-mounted camera, collect the surface images of traffic infrastructure, including the surface images of roads, bridges, and tunnels, while the vehicle is driving; S2. Set indicators for the operation status of transportation infrastructure; S3. For the surface images of the transportation infrastructure collected in step S1, select all images including the operating status indicators of the transportation infrastructure, annotate the operating status indicators by manual annotation, and then construct a data set for detecting and evaluating the operating status of urban transportation infrastructure; S4. Construct an intelligent identification model for traffic infrastructure operation status indicators, including a basic model, a first auxiliary model, a second auxiliary model, and an enhanced model; S5. construct the loss function among the output layers of the basic model, the first auxiliary model, the second auxiliary model, and the enhanced model, and then construct the overall loss function among the basic model, the auxiliary model, and the enhanced model; S6. Using the data set for urban transportation infrastructure operation status detection and evaluation obtained in step S3, the constructed basic model, the first auxiliary model, the second auxiliary model, and the enhanced model are trained to obtain the trained basic model as the final intelligent identification model of transportation infrastructure operation status indicators; S7. Constructing a digital twin model based on the intelligent identification model of the transportation infrastructure operation status indicator obtained in step S6; S8. Based on the digital twin model obtained in step S7, a method for evaluating the operating status of transportation infrastructure is constructed to evaluate the operating status of transportation infrastructure.

[0006] Furthermore, among the operating status indicators of the traffic infrastructure in step S2, the operating status indicators of roads include: cracks, rutting, waves, potholes, oil spills, repairs, and looseness; the operating status indicators of bridges include: cracks, spalling, steel corrosion, water seepage, honeycombs, rough surfaces, and demolding defects; the operating status indicators of tunnels include: cracks, spalling, steel corrosion, water seepage, and drainage system blockage.

[0007] Furthermore, the specific implementation method of step S3 includes the following steps: S3.1. Manually select the surface images of the transportation infrastructure collected in step S1, and then manually label the operating status indicators of roads, bridges, and tunnels; S3.2. The labeled traffic infrastructure appearance images are used as a dataset for urban traffic infrastructure operation status detection and evaluation, and are divided into a training set and a test set in a ratio of 7:3.

[0008] Furthermore, the specific implementation method of step S4 includes the following steps: S4.1. Basic model for constructing intelligent identification model of traffic infrastructure operation status indicators: S4.1.1. The basic model A consists of 1 input layer, 14 convolutional layers, 14 activation layers, 5 pooling layers, 1 feature splitting layer, 4 concatenation layers, 4 convolutional upsampling layers, and 1 output layer. S4.1.2. Input the training set obtained in step S3 into the basic model A, calculate the weights and biases of the neural network units through the error back propagation algorithm, complete the training of the basic model A, and obtain the trained basic model A; The calculation formula of the loss function of the basic model A is: ; in, is the loss function of the basic model A; is the total number of data samples in the basic model A; For the ia The true labels of samples, Based on the prediction of model A ia The probability that a sample is class 1; S4.1.3. The node output values ​​of the 19 output units corresponding to the basic model A are a1, a2, a3, a4, ..., a19. The node output values ​​of the 19 output units are initially converted to obtain the output values The corresponding probability value The calculation formula is: ; In order to amplify the probability difference between the output values ​​and enhance the model's ability to distinguish, the 19 node outputs are further optimized and converted into: ; in, G is the enhancement coefficient, and then the actual disease corresponding to each node is marked as ; S4.1.4. Adoption The training set obtained in step S3 is combined with the error chain derivation rule to train the basic model A. After the training is completed, the output values ​​of the 19 types of diseases corresponding to the basic model A are obtained, which are recorded as , , , , …, ; S4.2. Constructing the first auxiliary model of the intelligent identification model of traffic infrastructure operation status indicators: S4.2.1. The first auxiliary model B consists of 1 input layer, 18 convolutional layers, 18 activation layers, 5 pooling layers, 1 feature splitting layer, 4 concatenation layers, 4 convolutional upsampling layers, and 1 output layer; S4.2.2. Input the training set obtained in step S3 into the first auxiliary model B, calculate the weights and biases of the neural network units through the error back propagation algorithm, complete the training of the first auxiliary model B, and obtain the trained first auxiliary model B; The calculation formula of the loss function of the first auxiliary model B is: ; in, is the loss function of the first auxiliary model B; is the total number of data samples in the first auxiliary model B; For the ib The true labels of samples, The first auxiliary model B predicts ib The probability that a sample is class 1; S4.2.3. The node output values ​​of the 19 output units corresponding to the first auxiliary model B are b1, b2, b3, b4, ..., b19. The node output values ​​of the 19 output units are initially converted to obtain the output values b i The corresponding probability value The calculation formula is: ; Then the node output values ​​of the 19 output units are further optimized and converted to obtain the output value b i The corresponding enhancement probability value The calculation formula is: ; S4.2.4. Adoption The training set obtained in step S3 is combined with the error chain derivation rule to train the first auxiliary model B. After the training is completed, the disease output value corresponding to the first auxiliary model B is obtained as , among the 19 nodes in the output layer, the actual disease corresponding to each node is marked as , i =1,2,3,…,19; S4.3. Constructing the second auxiliary model of the intelligent identification model of traffic infrastructure operation status indicators: S4.3.1. The second auxiliary model C consists of 1 input layer, 22 convolutional layers, 22 activation layers, 6 pooling layers, 1 feature splitting layer, 5 concatenation layers, 5 convolution upsampling layers, and 1 output layer; S4.3.2. Input the training set obtained in step S3 into the second auxiliary model C, calculate the weights and biases of the neural network units through the error back propagation algorithm, complete the training of the second auxiliary model C, and obtain the trained second auxiliary model C; The calculation formula of the loss function of the second auxiliary model C is: ; in, is the loss function of the second auxiliary model C; is the total number of data samples in the second auxiliary model C; For the ic The true labels of samples, The second auxiliary model C predicts ic The probability that a sample is class 1; S4.3.3. The node output values ​​of the 19 output units corresponding to the second auxiliary model C are c1, c2, c3, c4, ..., c19. The node output values ​​of the 19 output units are initially converted to obtain The corresponding probability value The calculation formula is: ; Then the node output values ​​of the 19 output units are further optimized and converted to obtain the output value The corresponding enhancement probability value The calculation formula is: ; in, G is the enhancement coefficient, and then the actual disease corresponding to each node is marked as ; S4.3.4. Adoption The training set obtained in step S3 is combined with the error chain derivation rule to train the second auxiliary model C. After the training is completed, the output values ​​of the 19 types of diseases corresponding to the second auxiliary model C are obtained as follows: ; S4.4. Constructing an enhanced model for intelligent identification of traffic infrastructure operation status indicators: S4.4.1. The enhanced model D consists of 1 input layer, 26 convolutional layers, 26 activation layers, 7 pooling layers, 1 feature splitting layer, 6 concatenation layers, 6 convolutional upsampling layers, and 1 output layer. S4.4.2. Input the training set obtained in step S3 into the enhanced model D, calculate the weights and biases of the neural network units through the error back propagation algorithm, complete the training of the enhanced model D, and obtain the trained enhanced model D; The loss function of the enhanced model D is calculated as: ; in, is the loss function of the enhanced model D; is the total number of data samples in the reinforcement model D; For the id The true labels of samples, To strengthen the model D to predict the id The probability that a sample is class 1; S4.4.3. The node output values ​​of the 19 output units corresponding to the enhanced model D are d1, d2, d3, d4, ..., d19. The node output values ​​of the 19 output units are initially converted to obtain The corresponding probability value The calculation formula is: ; Then the node output values ​​of the 19 output units are further optimized and converted to obtain the output value The corresponding enhancement probability value The calculation formula is: ; in, G is the enhancement coefficient, and then the actual disease corresponding to each node is marked as ; S4.4.4. Adoption The training set obtained in step S3 is combined with the error chain derivation rule to train the enhanced model D. After the training is completed, the output values ​​of the 19 types of diseases corresponding to the enhanced model D are obtained as follows: .

[0009] Furthermore, the specific implementation method of step S5 includes the following steps: S5.1. Based on and , establish the loss function between the first auxiliary model B and the output layer of the basic model A, and the calculation formula is: ; based on and , construct the output value The corresponding probability value The loss function of the base model A , the calculation formula is: ; based on and , establish the relationship loss function between the output layer of the first auxiliary model B and the basic model A , the calculation formula is: ; in for The weight coefficient of for The weight coefficient of S5.2. Establish the relationship loss function between the output layer of the second auxiliary model C and the first auxiliary model B , the expression is: ; in, for The weight coefficient of for The weight coefficient of Based on and The loss function of Based on and The loss function of The calculation formula is: ; The calculation formula is: ; S5.3. Establish the relationship loss function between the output layer of the second auxiliary model C and the basic model A , the expression is: ; in, for The weight coefficient of for The weight coefficient of Based on and The loss function is Based on and The loss function of The calculation formula is: ; The calculation formula is: ; S5.4. Establishing the relationship loss function between the output layer of the enhanced model D and the base model A , the expression is: ; in, for The weight coefficient of for The weight coefficient of Based on and The loss function of The calculation formula is: ; The calculation formula is: ; S5.5. Establish the relationship loss function between the output layer of the reinforcement model D and the second auxiliary model C , the expression is: ; in, for The weight coefficient of for The weight coefficient of Based on and The loss function is Based on and The loss function of The calculation formula is: ; The calculation formula is: ; S5.6. Overall loss function among the constructed base model, first auxiliary model, second auxiliary model, and enhanced model , the calculation formula is: .

[0010] Furthermore, the specific implementation method of step S6 is to use the data set for urban traffic infrastructure operation status detection and evaluation obtained in step S3, based on the chain derivation rule, and train the basic model, the first auxiliary model, the second auxiliary model, and the reinforcement model constructed for rutting recognition through error back propagation. When the overall loss function obtained in step S5 When is minimum, the corresponding model parameters are optimal, and the trained basic model A is the final intelligent identification model for transportation infrastructure operation status indicators.

[0011] Furthermore, the specific implementation method of step S7 includes the following steps: S7.1. Export the intelligent identification model of traffic infrastructure operation status indicators obtained in step S6 as a deployment file for independent services, including one of a pt file, a pth file, a json file, and a txt file; S7.2. Build a transportation infrastructure information model; S7.2.1. Collect information on the location, length, width, and layout of transportation infrastructure; S7.2.2. The data obtained in step S7.2.1 is stored in a database, and a basic three-dimensional model of the transportation infrastructure is constructed, wherein the basic three-dimensional model of the transportation infrastructure includes geometric information of the transportation infrastructure; S7.2.3. Based on step S7.2.2, add material properties, design specifications, and historical maintenance data to the basic three-dimensional model of the transportation infrastructure to obtain a transportation infrastructure information model; S7.2.4. Export the transportation infrastructure information model to a format file supported by the web application, including a json file or a js file; S7.3. In the digital twin platform, import the deployment file of the intelligent identification model of the traffic infrastructure operation status indicator obtained in step S7.1, and import the traffic infrastructure information model format file obtained in step S7.2; S7.4. Use Python to construct an API interface between image data and the intelligent identification model of traffic infrastructure operation status indicators for intelligent identification of ruts; use Python to construct an API interface between the intelligent identification model of traffic infrastructure operation status indicators and the traffic infrastructure information model to complete the construction of the digital twin model.

[0012] Furthermore, the specific implementation method of step S8 includes the following steps: S8.1. Use a camera to collect road surface data and upload it to the digital twin platform. Then use the digital twin model to intelligently identify various indicators of transportation infrastructure in the image, select all images that include indicators of transportation infrastructure operation status, conduct on-site review of the indicators, and measure the indicators to obtain data that includes all indicators of transportation infrastructure operation status; S8.2. Propose a method for evaluating the operating status of facilities on roads; S8.2.1. Set based on crack length LCD , crack width LKD , rutting area CMJ , wave area BMJ , pit area KMJ , Oil spill area FMJ , Repair area XMJ , loose area SMJ , establish road operation status evaluation standards and divide the road operation status standards into 5 levels; S8.2.2. For each level of road operating condition standard in step S8.2.1, establish a road operating condition standard matrix. : , na The road operation status standard is graded as 1, 2, 3, 4, and 5. , , , , , , , They are the standard values ​​of the grades corresponding to crack length, crack width, rutting area, wave area, pothole area, oil spill area, repair area, and loose area; S8.2.3. Based on the collected images of the target road, a rapid intelligent identification method for the operating status indicators of urban transportation infrastructure is used to identify the yx Cracks, ruts, waves, potholes, oil spills, repairs, loose diseases of the road, and measure and calculate the yx Length of cracks on roads , crack width , rutting area , wave area , pit area , Oil spill area , Repair area , loose area , and establish the road operation status matrix : ; S8.2.4. Comparison Matrix with 5 road operating conditions standards The standard value in is used to score each indicator value by expert scoring to construct a classification vector, where the expert scoring is divided into 1~5 points, including 1.5, 2.5, 3.5, and 4.5, where 1 means that the current indicator status is recommended to be divided into Class 1, 1.5 means that the current indicator status is recommended to be divided between Class 1 and Class 2, 2 means that the current indicator status is recommended to be divided into Class 2, 2.5 means that the current indicator status is recommended to be divided between Class 2 and Class 3, 3 means that the current indicator status is recommended to be divided into Class 3, 3.5 means that the current indicator status is recommended to be divided between Class 3 and Class 4, 4 means that the current indicator status is recommended to be divided into Class 4, 4.5 means that the current indicator status is recommended to be divided between Class 4 and Class 5, and 5 means that the current indicator status is recommended to be divided into Class 5; The road operation status scoring matrix is: ; in, ~ The scores represent the length of road cracks, the width of road cracks, the rutting area, the wave area, the pothole area, the oil spill area, the repair area, and the loose area; Based on the above scoring values, calculate ~ The weighted average score , the calculation formula is: ; in, Indicates the weight value corresponding to all road indicators; Compare The relationship between 1,2,3,4,5 and The closest value is the road operation status classification result, which is recorded as DL ; S8.3. Propose a method for evaluating the operating condition of facilities for bridges; S8.3.1. Set the length of cracks based on the bridge surface. QCD , crack width QKD , peeling area QBL , Steel bar corrosion area QXS , Seepage area QSS , Cell area QFW , pockmarked surface area QMM , demoulding defect area QTM , establish the bridge operation status evaluation standard and divide the bridge operation status standard into 5 levels; S8.3.2. For each level of bridge performance criteria, establish a bridge performance criteria matrix. , , nb The grade numbers for the bridge operation status standards are 1, 2, 3, 4, and 5; , , , , , , , They are the standard values ​​of the grades corresponding to the crack length, crack width, peeling area, steel bar corrosion area, water seepage area, honeycomb area, pitting area, and demoulding defect area; S8.3.3. Based on the collected images of the target bridge, a rapid intelligent identification method for the operating status indicators of urban transportation infrastructure is used to identify the yx bridge cracks, spalling, steel bar corrosion, water seepage, honeycomb, pockmarks, demoulding defects, and measure and calculate the yx Length of cracks in bridges , crack width , peeling area , Steel bar corrosion area , Seepage area , Cell area , pockmarked surface area , demoulding defect area , and establish the yx Bridge operation status matrix : ; S8.3.4. Comparison Matrix with 5 bridge operating status standards The standard value in is used to score each indicator value by expert scoring to construct a classification vector, where the expert scoring is divided into 1~5 points, including 1.5, 2.5, 3.5, and 4.5, where 1 means that the current indicator status is recommended to be divided into Class 1, 1.5 means that the current indicator status is recommended to be divided between Class 1 and Class 2, 2 means that the current indicator status is recommended to be divided into Class 2, 2.5 means that the current indicator status is recommended to be divided between Class 2 and Class 3, 3 means that the current indicator status is recommended to be divided into Class 3, 3.5 means that the current indicator status is recommended to be divided between Class 3 and Class 4, 4 means that the current indicator status is recommended to be divided into Class 4, 4.5 means that the current indicator status is recommended to be divided between Class 4 and Class 5, and 5 means that the current indicator status is recommended to be divided into Class 5; The bridge operation status scoring matrix is: ; in, ~ They represent the scores of bridge crack length, bridge crack width, peeling area, steel bar corrosion area, water seepage area, honeycomb area, pitting area, and demoulding defect area; Based on the above scoring values, calculate ~ The weighted average score , the calculation formula is: ; in, Indicates the weight values ​​corresponding to all indicators of the bridge; comparison The relationship between 1,2,3,4,5 and The closest value is the classification result of the bridge operation status QL ; S8.4. Propose a method for evaluating the operating status of facilities in tunnels; S8.4.1. Set the crack length based on the tunnel surface SCD , crack width SKD , peeling area SBL , Steel bar corrosion area SXS , Seepage area SSS 、Number of drainage system blockages SSD , establish tunnel operation status evaluation standards and divide the tunnel operation status standards into 5 levels; S8.4.2. For each level of tunnel operating condition standard, establish a tunnel operating condition standard matrix. , , nc The level numbers for the tunnel operation status standards are 1, 2, 3, 4, and 5; , , , , , They are the standard values ​​corresponding to crack length, crack width, peeling area, steel bar corrosion area, water seepage area, and number of drainage system blockages; S8.4.3. Comparison Matrix with 5 tunnel operation status standards The standard value in is used to score each indicator value by expert scoring to construct a classification vector; the expert scoring is divided into 1~5 points, including 1.5, 2.5, 3.5, and 4.5, where 1 means that the current indicator status is recommended to be classified as category 1, 1.5 means that the current indicator status is recommended to be classified between category 1 and category 2, 2 means that the current indicator status is recommended to be classified as category 2, 2.5 means that the current indicator status is recommended to be classified between category 2 and category 3, 3 means that the current indicator status is recommended to be classified as category 3, 3.5 means that the current indicator status is recommended to be classified between category 3 and category 4, 4 means that the current indicator status is recommended to be classified as category 4, 4.5 means that the current indicator status is recommended to be classified between category 4 and category 5, and 5 means that the current indicator status is recommended to be classified as category 5; The tunnel operation status scoring matrix is: ; in, ~ The scores represent the length of tunnel cracks, the width of tunnel cracks, the spalling area, the steel bar corrosion area, the water seepage area, and the number of drainage system blockages; Based on the above scoring values, calculate ~ The weighted average score , the calculation formula is: ; in, Indicates the weight value corresponding to all tunnel indicators; Compare The relationship between 1,2,3,4,5 and The closest value is the tunnel operation status classification result SD ; S8.5. Based on the above calculation DL , QL , SD , calculate the comprehensive evaluation level of transportation infrastructure operation status , the calculation formula is: ; in, is the weight of the road operation status classification result, is the weight of the bridge operation status classification result, is the weight of the tunnel operation status classification result; At this point, the evaluation of the operating status of transportation infrastructure is completed.

[0013] Beneficial effects of the present invention: The method for detecting and evaluating the operation status of urban transportation infrastructure described in the present invention uses the same equipment to simultaneously collect data on facilities such as roads, bridges, and tunnels, thereby increasing detection speed and efficiency and reducing detection costs. At the same time, the introduction of intelligent methods can greatly improve the accuracy and reliability of facility detection and evaluation, improve the timeliness of decision-making, and improve the speed and accuracy of detection of indicators of the operation status of transportation infrastructure.

[0014] The method for detecting and evaluating the operation status of urban transportation infrastructure described in the present invention, in the intelligent identification model of the operation status index of the transportation infrastructure constructed, the input layer is responsible for receiving the original input transportation infrastructure image data; the convolution layer is used to scan the local area of ​​the image through the convolution kernel to extract low-level to high-level features. The low-level features include the edges, textures and other features of the collected images of the transportation infrastructure, and the high-level features include the overall features of the operation status indicators of the transportation infrastructure. By sharing parameters, the model parameters are significantly reduced and the calculation efficiency is improved. The feature splitting layer is used to decompose and reorganize the extracted operation status indicator features of the transportation infrastructure, so that the expression form of the features is clearer and independent, which is convenient for the processing of the subsequent layers. The activation layer is used to perform nonlinear transformation on the linear combination of the output of the convolution layer, increase the expression ability of the model, and enable it to learn more complex operation status indicator features of the transportation infrastructure. The pooling layer is used to reduce the size of the feature map and reduce the amount of calculation through downsampling operations, while retaining important features and enhancing the noise resistance of the model. The splicing layer is used to splice or merge the features from different convolutional layers or pooling layers, integrate the feature information at different levels, and enhance the expression ability of the operation status indicator identification model of the transportation infrastructure. Convolution upsampling layer: restores the feature map to the same resolution as the input image, generates fine segmentation results, and improves the accuracy of feature extraction of traffic infrastructure operation status indicators. Output layer: generates the final traffic infrastructure operation status indicator recognition results based on the features extracted and processed by the previous layers. The learning objectives of different models are linked through loss functions, and collaborative training improves the generalization ability of the model, which helps to extract more robust features and improve the speed and accuracy of detection and recognition of urban traffic infrastructure operation status indicators.

[0015] The method for detecting and evaluating the operating status of urban transportation infrastructure described in the present invention realizes efficient and accurate evaluation of the operating status of facilities. It can evaluate the status and comprehensive evaluation of roads, bridges, and tunnels, improve maintenance efficiency and safety, and bring significant improvements and optimizations to the management and maintenance of transportation infrastructure. BRIEF DESCRIPTION OF THE DRAWINGS

[0016] Figure 1 The present invention is a flow chart of a method for detecting and evaluating the operating status of urban transportation infrastructure. DETAILED DESCRIPTION

[0017] In order to make the purpose, technical solution and advantages of the present invention more clear, the present invention is further described in detail below in conjunction with the accompanying drawings and specific embodiments. It should be understood that the specific embodiments described herein are only used to explain the present invention and are not used to limit the present invention, that is, the specific embodiments described are only part of the embodiments of the present invention, rather than all of the specific embodiments. The components of the specific embodiments of the present invention described and shown in the drawings herein can be arranged and designed in various different configurations, and the present invention can also have other embodiments.

[0018] Therefore, the following detailed description of the specific embodiments of the present invention provided in the accompanying drawings is not intended to limit the scope of the invention claimed for protection, but merely represents the selected specific embodiments of the present invention. Based on the specific embodiments of the present invention, all other specific embodiments obtained by those skilled in the art without making creative work are within the scope of protection of the present invention.

[0019] In order to further understand the content, features and effects of the present invention, the following specific implementation methods are given as examples, and the attached Figure 1 The detailed instructions are as follows:

[0020] Embodiment 1: A method for detecting and evaluating the operating status of urban transportation infrastructure comprises the following steps: S1. Using a vehicle-mounted camera, collect the surface images of traffic infrastructure, including the surface images of roads, bridges, and tunnels, while the vehicle is driving; S2. Set indicators for the operation status of transportation infrastructure; In the traffic infrastructure operation status indicators in step S2, the road operation status indicators include: cracks, rutting, waves, potholes, oil spills, repairs, and looseness; the bridge operation status indicators include: cracks, spalling, steel corrosion, water seepage, honeycomb, pitting, and demoulding defects; the tunnel operation status indicators include: cracks, spalling, steel corrosion, water seepage, and drainage system blockage; S3. For the surface images of the transportation infrastructure collected in step S1, select all images including the operating status indicators of the transportation infrastructure, annotate the operating status indicators by manual annotation, and then construct a data set for detecting and evaluating the operating status of urban transportation infrastructure; Furthermore, the specific implementation method of step S3 includes the following steps: S3.1. Manually select the surface images of the transportation infrastructure collected in step S1, and then manually label the operating status indicators of roads, bridges, and tunnels; S3.2. The labeled traffic infrastructure appearance images are used as a dataset for urban traffic infrastructure operation status detection and evaluation, and are divided into a training set and a test set in a ratio of 7:3; S4. Construct an intelligent identification model for traffic infrastructure operation status indicators, including a basic model, a first auxiliary model, a second auxiliary model, and an enhanced model; Furthermore, the specific implementation method of step S4 includes the following steps: S4.1. Basic model for constructing intelligent identification model of traffic infrastructure operation status indicators: S4.1.1. The basic model A consists of 1 input layer, 14 convolutional layers, 14 activation layers, 5 pooling layers, 1 feature splitting layer, 4 concatenation layers, 4 convolution upsampling layers, and 1 output layer. The specific network structure is the following structure connected in sequence: Input layer: image size 227×227×3; Convolution layer 1: convolution kernel size 3×3; number of convolution kernels 32; stride: 1; Activation layer 1: The activation function is LeakyReLU; Convolutional layer 2: convolution kernel size 3×3; number of convolution kernels 32; Stride: 1; Activation layer 2: The activation function is LeakyReLU; Pooling layer 1: Type is maximum pooling; pooling size: 2×2; Stride: 2; Convolution layer 3: convolution kernel size 3×3; number of convolution kernels 64; stride: 1; Activation layer 3: The activation function is LeakyReLU; Convolutional layer 4: convolution kernel size 3×3; number of convolution kernels 64; Stride: 1; Activation layer 4: The activation function is LeakyReLU; Pooling layer 2: Type is maximum pooling; pooling size: 2×2; Stride: 2; Convolution layer 5: convolution kernel size 3×3; number of convolution kernels 128; stride: 1; Activation layer 5: The activation function is LeakyReLU; Convolution layer 6: convolution kernel size 3×3; number of convolution kernels 128; stride: 1; Activation layer 6: The activation function is LeakyReLU; Pooling layer 3: Type is maximum pooling; pooling size: 2×2; Stride: 2; Convolution layer 7: convolution kernel size 3×3; number of convolution kernels 256; stride: 1; Activation layer 7: The activation function is LeakyReLU; Convolution layer 8: convolution kernel size 3×3; number of convolution kernels 256; stride: 1; Activation layer 8: The activation function is LeakyReLU; Pooling layer 4: Type is maximum pooling; pooling size: 2×2; Stride: 2; Convolution layer 9: convolution kernel size 3×3; number of convolution kernels 512; stride: 1; Activation layer 9: The activation function is LeakyReLU; Convolution layer 10: kernel size 3×3; number of kernels 512; stride: 1; Activation layer 10: The activation function is LeakyReLU; Pooling layer 5: Type is maximum pooling; pooling size: 2×2; Stride: 2; Feature splitting layer: Split the 512-dimensional vector obtained above into two 256-dimensional vectors, where For the first 256-dimensional vector, perform the following operations: 1. Convolution layer 1-1: convolution kernel size 3×3; number of convolution kernels 256; stride: 1; 2. Convolutional layer 1-2: convolution kernel size 3×3; number of convolution kernels 256; stride: 1; 3. Convolutional layers 1-3: kernel size 3×3; number of kernels 256; Stride: 1; 4. Convolutional layers 1-4: kernel size 3×3; number of kernels 256; Stride: 1; 5. Feature addition layer: Add the output vectors of convolutional layers 1-2 and 1-4 to obtain a 256-dimensional vector; 6. Feature connection layer: connect the first 256-dimensional vector, the output of convolutional layers 1-2, and the data of the feature addition layer through channels, and output a 768-dimensional vector; For the second 256-dimensional vector, perform the following operations: 1. Perform channel connection with the output of the sixth feature connection layer in the first 256-dimensional vector to output a 1024-dimensional vector; 2. Convolutional layer 2-1: convolution kernel size 3×3; number of convolution kernels 512; Stride: 1; Convolution upsampling 1: Use 2×2 transposed convolution upsampling, and the output channels are halved to 256; Concatenation layer 1: connects the feature map with the output of activation layer 8, and the number of channels increases to 512; Convolution layer 11: convolution kernel size 3×3; number of convolution kernels 256; stride: 1; Activation layer 11: The activation function is LeakyReLU; Convolution upsampling 2: Use 2×2 transposed convolution upsampling, and the output channels are halved to 128; Concatenation layer 2: connects the feature map with the output of activation layer 6, and the number of channels increases to 256; Convolution layer 12: convolution kernel size 3×3; number of convolution kernels 128; stride: 1; Activation layer 12: The activation function is LeakyReLU; Convolution upsampling 3: Use 2×2 transposed convolution upsampling, and the output channels are halved to 64; Concatenation layer 3: connects the feature map with the output of activation layer 4, and the number of channels increases to 128; Convolution layer 13: kernel size 3×3; number of kernels 64; stride: 1; Activation layer 13: The activation function is LeakyReLU; Convolution upsampling 4: Use 2×2 transposed convolution upsampling, and the output channels are halved to 32; Concatenation layer 4: connects the feature map with the output of activation layer 2, and the number of channels increases to 64; Convolution layer 14: kernel size 3×3; number of kernels 32; stride: 1; Activation layer 14: the activation function is LeakyReLU; Output layer: type is normalized exponential function Softmax; number of neurons: 19; S4.1.2. Input the training set obtained in step S3 into the basic model A, calculate the weights and biases of the neural network units through the error back propagation algorithm, complete the training of the basic model A, and obtain the trained basic model A; The calculation formula of the loss function of the basic model A is: ; in, is the loss function of the basic model A; is the total number of data samples in the basic model A; For the ia The true labels of samples, Based on the prediction of model A iaThe probability that a sample is class 1; S4.1.3. The node output values ​​of the 19 output units corresponding to the basic model A are a1, a2, a3, a4, ..., a19. The node output values ​​of the 19 output units are initially converted to obtain the output values The corresponding probability value The calculation formula is: ; In order to amplify the probability difference between the output values ​​and enhance the model's ability to distinguish, the 19 node outputs are further optimized and converted into: ; in, G is the enhancement coefficient, and then the actual disease corresponding to each node is marked as ; Furthermore, cracks, rutting, waves, potholes, oil spills, repairs, looseness of roads, cracks, spalling, steel corrosion, water seepage, honeycombs, pitting, demoulding defects of bridges, cracks, spalling, steel corrosion, water seepage, and drainage system blockage of tunnels, corresponding =1~19; S4.1.4. Adoption The training set obtained in step S3 is combined with the error chain derivation rule to train the basic model A. After the training is completed, the output values ​​of the 19 types of diseases corresponding to the basic model A are obtained, which are recorded as , , , , …, ; S4.2. Constructing the first auxiliary model of the intelligent identification model of traffic infrastructure operation status indicators: S4.2.1. The first auxiliary model B consists of 1 input layer, 18 convolutional layers, 18 activation layers, 5 pooling layers, 1 feature splitting layer, 4 concatenation layers, 4 convolution upsampling layers, and 1 output layer. The specific network structure is the following structure connected in sequence: Input layer: image size 227×227×3; Convolution layer 1: convolution kernel size 3×3; number of convolution kernels 64; Stride: 1; Activation layer 1: The activation function is LeakyReLU; Convolutional layer 2: convolution kernel size 3×3; number of convolution kernels 64; stride: 1; Activation layer 2: The activation function is LeakyReLU; Pooling layer 1: Type is maximum pooling; pooling size: 2×2; Stride: 2; Convolution layer 3: convolution kernel size 3×3; number of convolution kernels 128; stride: 1; Activation layer 3: The activation function is LeakyReLU; Convolution layer 4: convolution kernel size 3×3; number of convolution kernels 128; stride: 1; Activation layer 4: The activation function is LeakyReLU; Pooling layer 2: Type is maximum pooling; pooling size: 2×2; Stride: 2; Convolution layer 5: convolution kernel size 3×3; number of convolution kernels 256; stride: 1; Activation layer 5: The activation function is LeakyReLU; Convolution layer 6: convolution kernel size 3×3; number of convolution kernels 256; stride: 1; Activation layer 6: The activation function is LeakyReLU; Pooling layer 3: Type is maximum pooling; pooling size: 2×2; Stride: 2; Convolution layer 7: kernel size 3×3; number of kernels 512; stride: 1; Activation layer 7: The activation function is LeakyReLU; Convolution layer 8: convolution kernel size 3×3; number of convolution kernels 512; stride: 1; Activation layer 8: The activation function is LeakyReLU; Pooling layer 4: Type is maximum pooling; pooling size: 2×2; Stride: 2; Convolution layer 9: convolution kernel size 3×3; number of convolution kernels 1024; stride: 1; Activation layer 9: The activation function is LeakyReLU; Convolution layer 10: kernel size 3×3; number of kernels 1024; stride: 1; Activation layer 10: The activation function is LeakyReLU; Pooling layer 5: Type is maximum pooling; pooling size: 2×2; Stride: 2; Feature splitting layer: Split the 1024-dimensional vector obtained above into two 512-dimensional vectors, where For the first 512-dimensional vector, perform the following operations: 1. Convolution layer 1-1: convolution kernel size 3×3; number of convolution kernels 512; stride: 1; 2. Convolutional layer 1-2: convolution kernel size 3×3; number of convolution kernels 512; stride: 1; 3. Convolutional layers 1-3: kernel size 3×3; number of kernels 512; Stride: 1; 4. Convolutional layers 1-4: kernel size 3×3; number of kernels 512; Stride: 1; 5. Feature addition layer: Add the output vectors of convolutional layers 1-2 and 1-4 to obtain a 512-dimensional vector; 6. Feature connection layer: connect the first 512-dimensional vector, the output of convolution layer 1-2, and the data of feature addition layer through channels, and output a 1536-dimensional vector; For the second 512-dimensional vector, perform the following operations: 1. Perform channel connection with the output of the feature connection layer in step 6 above and output a 2048-dimensional vector; 2. Convolutional layer 2-1: convolution kernel size 3×3; number of convolution kernels 1024; stride: 1; Convolution upsampling 1: Use 2×2 transposed convolution upsampling, and the output channels are halved to 512; Concatenation layer 1: connects the feature map with the output of activation layer 8, and the number of channels increases to 1024; Convolution layer 11: convolution kernel size 3×3; number of convolution kernels 512; stride: 1; Activation layer 11: The activation function is LeakyReLU; Convolution layer 12: convolution kernel size 3×3; number of convolution kernels 512; stride: 1; Activation layer 12: LeakyReLU; Convolution upsampling 2: Use 2×2 transposed convolution upsampling, and the output channels are halved to 256; Concatenation layer 2: connects the feature map with the output of activation layer 6, and the number of channels increases to 512; Convolution layer 13: convolution kernel size 3×3; number of convolution kernels 256; stride: 1; Activation layer 13: The activation function is LeakyReLU; Convolution layer 14: kernel size 3×3; number of kernels 256; stride: 1; Activation layer 14: the activation function is LeakyReLU; Convolution upsampling 3: Use 2×2 transposed convolution upsampling, and the output channels are halved to 128; Concatenation layer 3: connects the feature map with the output of activation layer 4, and the number of channels increases to 256; Convolution layer 15: kernel size 3×3; number of kernels 128; stride: 1; Activation layer 15: The activation function is LeakyReLU; Convolution layer 16: kernel size 3×3; number of kernels 128; stride: 1; Activation layer 16: The activation function is LeakyReLU; Convolution upsampling 4: Use 2×2 transposed convolution upsampling, and the output channels are halved to 64; Concatenation layer 4: connects the feature map with the output of activation layer 2, and the number of channels increases to 128; Convolution layer 17: kernel size 3×3; number of kernels 64; stride: 1; Activation layer 17: The activation function is LeakyReLU; Convolution layer 18: convolution kernel size 3×3; number of convolution kernels 64; stride: 1; Activation layer 18: The activation function is LeakyReLU; Output layer: type is normalized exponential function Softmax; number of neurons: 19; S4.2.2. Input the training set obtained in step S3 into the first auxiliary model B, calculate the weights and biases of the neural network units through the error back propagation algorithm, complete the training of the first auxiliary model B, and obtain the trained first auxiliary model B; The calculation formula of the loss function of the first auxiliary model B is: ; in, is the loss function of the first auxiliary model B; is the total number of data samples in the first auxiliary model B; For the ib The true labels of samples, The first auxiliary model B predicts ib The probability that a sample is class 1; S4.2.3. The node output values ​​of the 19 output units corresponding to the first auxiliary model B are b1, b2, b3, b4, ..., b19. The node output values ​​of the 19 output units are initially converted to obtain the output values b i The corresponding probability value The calculation formula is: ; Then the node output values ​​of the 19 output units are further optimized and converted to obtain the output value b i The corresponding enhancement probability value The calculation formula is: ; S4.2.4. Adoption The training set obtained in step S3 is combined with the error chain derivation rule to train the first auxiliary model B. After the training is completed, the disease output value corresponding to the first auxiliary model B is obtained as , among the 19 nodes in the output layer, the actual disease corresponding to each node is marked as , i =1,2,3,…,19; Furthermore, cracks, rutting, waves, potholes, oil spills, repairs, looseness of roads, cracks, spalling, steel corrosion, water seepage, honeycombs, pitting, demoulding defects of bridges, cracks, spalling, steel corrosion, water seepage, and drainage system blockage of tunnels, corresponding =1~19; S4.3. Constructing the second auxiliary model of the intelligent identification model of traffic infrastructure operation status indicators: S4.3.1. The second auxiliary model C consists of 1 input layer, 22 convolutional layers, 22 activation layers, 6 pooling layers, 1 feature splitting layer, 5 concatenation layers, 5 convolution upsampling layers, and 1 output layer. The specific network structure is the following structure connected in sequence: Input layer: image size 227×227×3; Convolution layer 1: convolution kernel size 3×3; number of convolution kernels 32; Stride: 1; Activation layer 1: The activation function is LeakyReLU; Convolutional layer 2: convolution kernel size 3×3; number of convolution kernels 32; Stride: 1; Activation layer 2: The activation function is LeakyReLU; Pooling layer 1: Type is maximum pooling; pooling size: 2×2; Stride: 2; Convolution layer 3: convolution kernel size 3×3; number of convolution kernels 64; stride: 1; Activation layer 3: The activation function is LeakyReLU; Convolutional layer 4: convolution kernel size 3×3; number of convolution kernels 64; Stride: 1; Activation layer 4: The activation function is LeakyReLU; Pooling layer 2: Type is maximum pooling; pooling size: 2×2; Stride: 2; Convolution layer 5: convolution kernel size 3×3; number of convolution kernels 128; stride: 1; Activation layer 5: The activation function is LeakyReLU; Convolution layer 6: convolution kernel size 3×3; number of convolution kernels 128; stride: 1; Activation layer 6: The activation function is LeakyReLU; Pooling layer 3: Type is maximum pooling; pooling size: 2×2; Stride: 2; Convolution layer 7: convolution kernel size 3×3; number of convolution kernels 256; stride: 1; Activation layer 7: The activation function is LeakyReLU; Convolution layer 8: convolution kernel size 3×3; number of convolution kernels 256; stride: 1; Activation layer 8: The activation function is LeakyReLU; Pooling layer 4: Type is maximum pooling; pooling size: 2×2; Stride: 2; Convolution layer 9: convolution kernel size 3×3; number of convolution kernels 512; stride: 1; Activation layer 9: The activation function is LeakyReLU; Convolution layer 10: kernel size 3×3; number of kernels 512; stride: 1; Activation layer 10: The activation function is LeakyReLU; Pooling layer 5: Type is maximum pooling; pooling size: 2×2; Stride: 2; Convolution layer 11: convolution kernel size 3×3; number of convolution kernels 1024; stride: 1; Activation layer 11: The activation function is LeakyReLU; Convolution layer 12: kernel size 3×3; number of kernels 1024; stride: 1; Activation layer 12: The activation function is LeakyReLU; Pooling layer 6: Type is maximum pooling; pooling size: 2×2; Stride: 2; Feature splitting layer: Split the 1024-dimensional vector obtained above into two 512-dimensional vectors, where For the first 512-dimensional vector, perform the following operations: 1. Convolution layer 1-1: convolution kernel size 3×3; number of convolution kernels 512; stride: 1; 2. Convolutional layer 1-2: convolution kernel size 3×3; number of convolution kernels 512; stride: 1; 3. Convolutional layers 1-3: kernel size 3×3; number of kernels 512; Stride: 1; 4. Convolutional layers 1-4: kernel size 3×3; number of kernels 512; Stride: 1; 5. Feature addition layer: Add the output vectors of convolutional layers 1-2 and 1-4 to obtain a 512-dimensional vector; 6. Feature connection layer: connect the first 512-dimensional vector, the output of convolution layer 1-2, and the data of feature addition layer through channels, and output a 1536-dimensional vector; For the second 512-dimensional vector, perform the following operations: 1. Perform channel connection with the output of the feature connection layer in step 6 above and output a 2048-dimensional vector; 2. Convolutional layer 2-1: convolution kernel size 3×3; number of convolution kernels 1024; stride: 1; Convolution upsampling 1: Use 2×2 transposed convolution upsampling, and the output channels are halved to 512; Concatenation layer 1: connects the feature map with the output of activation layer 10, and the number of channels increases to 1024; Convolution layer 13: kernel size 3×3; number of kernels 512; stride: 1; Activation layer 13: The activation function is LeakyReLU; Convolution layer 14: kernel size 3×3; number of kernels 512; stride: 1; Activation layer 14: the activation function is LeakyReLU; Convolution upsampling 2: Use 2×2 transposed convolution upsampling, and the output channels are halved to 256; Concatenation layer 2: connects the feature map with the output of activation layer 8, and the number of channels increases to 512; Convolution layer 15: kernel size 3×3; number of kernels 256; stride: 1; Activation layer 15: The activation function is LeakyReLU; Convolution layer 16: kernel size 3×3; number of kernels 256; stride: 1; Activation layer 16: The activation function is LeakyReLU; Convolution upsampling 3: Use 2×2 transposed convolution upsampling, and the output channels are halved to 128; Concatenation layer 3: connects the feature map with the output of activation layer 6, and the number of channels increases to 256; Convolution layer 17: kernel size 3×3; number of kernels 128; stride: 1; Activation layer 17: The activation function is LeakyReLU; Convolution layer 18: convolution kernel size 3×3; number of convolution kernels 128; stride: 1; Activation layer 18: The activation function is LeakyReLU; Convolution upsampling 4: Use 2×2 transposed convolution upsampling, and the output channels are halved to 64; Concatenation layer 4: connects the feature map with the output of activation layer 4, and the number of channels increases to 128; Convolution layer 19: convolution kernel size 3×3; number of convolution kernels 64; stride: 1; Activation layer 19: The activation function is LeakyReLU; Convolutional layer 20: convolution kernel size 3×3; number of convolution kernels 64; stride: 1; Activation layer 20: The activation function is LeakyReLU; Convolution upsampling 5: Use 2×2 transposed convolution upsampling, and the output channels are halved to 32; Concatenation layer 5: connects the feature map with the output of activation layer 2, and the number of channels increases to 64; Convolution layer 21: convolution kernel size 3×3; number of convolution kernels 32; stride: 1; Activation layer 21: The activation function is LeakyReLU; Convolutional layer 22: convolution kernel size 3×3; number of convolution kernels 32; stride: 1; Activation layer 22: The activation function is LeakyReLU; Output layer: Type: Softmax; Number of neurons: 19; S4.3.2. Input the training set obtained in step S3 into the second auxiliary model C, calculate the weights and biases of the neural network units through the error back propagation algorithm, complete the training of the second auxiliary model C, and obtain the trained second auxiliary model C; The calculation formula of the loss function of the second auxiliary model C is: ; in, is the loss function of the second auxiliary model C; is the total number of data samples in the second auxiliary model C; For the ic The true labels of samples, The second auxiliary model C predicts ic The probability that a sample is class 1; S4.3.3. The node output values ​​of the 19 output units corresponding to the second auxiliary model C are c1, c2, c3, c4, ..., c19. The node output values ​​of the 19 output units are initially converted to obtain The corresponding probability value The calculation formula is: ; Then the node output values ​​of the 19 output units are further optimized and converted to obtain the output value The corresponding enhancement probability value The calculation formula is: ; in, G is the enhancement coefficient, and then the actual disease corresponding to each node is marked as ; Furthermore, cracks, rutting, waves, potholes, oil spills, repairs, looseness of roads, cracks, spalling, steel corrosion, water seepage, honeycombs, pitting, demoulding defects of bridges, cracks, spalling, steel corrosion, water seepage, and drainage system blockage of tunnels, corresponding =1~19; S4.3.4. Adoption The training set obtained in step S3 is combined with the error chain derivation rule to train the second auxiliary model C. After the training is completed, the output values ​​of the 19 types of diseases corresponding to the second auxiliary model C are obtained as follows: ; S4.4. Constructing an enhanced model for intelligent identification of traffic infrastructure operation status indicators: S4.4.1. The enhanced model D consists of 1 input layer, 26 convolutional layers, 26 activation layers, 7 pooling layers, 1 feature splitting layer, 6 concatenation layers, 6 convolution upsampling layers, and 1 output layer. The specific network structure is the following structure connected in sequence: Input layer: image size 227×227×3; Convolution layer 1: convolution kernel size 3×3; number of convolution kernels 32; Stride: 1; Activation layer 1: The activation function is LeakyReLU; Convolutional layer 2: convolution kernel size 3×3; number of convolution kernels 32; Stride: 1; Activation layer 2: The activation function is LeakyReLU; Pooling layer 1: Type is maximum pooling; pooling size: 2×2; Stride: 2; Convolution layer 3: convolution kernel size 3×3; number of convolution kernels 64; stride: 1; Activation layer 3: The activation function is LeakyReLU; Convolutional layer 4: convolution kernel size 3×3; number of convolution kernels 64; Stride: 1; Activation layer 4: The activation function is LeakyReLU; Pooling layer 2: Type is maximum pooling; pooling size: 2×2; Stride: 2; Convolution layer 5: convolution kernel size 3×3; number of convolution kernels 128; stride: 1; Activation layer 5: The activation function is LeakyReLU; Convolution layer 6: convolution kernel size 3×3; number of convolution kernels 128; stride: 1; Activation layer 6: The activation function is LeakyReLU; Pooling layer 3: Type is maximum pooling; pooling size: 2×2; Stride: 2; Convolution layer 7: convolution kernel size 3×3; number of convolution kernels 256; stride: 1; Activation layer 7: The activation function is LeakyReLU; Convolution layer 8: convolution kernel size 3×3; number of convolution kernels 256; stride: 1; Activation layer 8: The activation function is LeakyReLU; Pooling layer 4: Type is maximum pooling; pooling size: 2×2; Stride: 2; Convolution layer 9: convolution kernel size 3×3; number of convolution kernels 512; stride: 1; Activation layer 9: The activation function is LeakyReLU; Convolution layer 10: kernel size 3×3; number of kernels 512; stride: 1; Activation layer 10: The activation function is LeakyReLU; Pooling layer 5: Type is maximum pooling; pooling size: 2×2; Stride: 2; Convolution layer 11: convolution kernel size 3×3; number of convolution kernels 1024; stride: 1; Activation layer 11: The activation function is LeakyReLU; Convolution layer 12: kernel size 3×3; number of kernels 1024; stride: 1; Activation layer 12: The activation function is LeakyReLU; Pooling layer 6: Type is maximum pooling; pooling size: 2×2; Stride: 2; Convolution layer 13: kernel size 3×3; number of kernels 2048; stride: 1; Activation layer 13: The activation function is LeakyReLU; Convolution layer 14: kernel size 3×3; number of kernels 2048; stride: 1; Activation layer 14: the activation function is LeakyReLU; Pooling layer 7: Type is max pooling; pooling size: 2×2; stride: 2; Feature splitting layer: Split the 2048-dimensional vector obtained above into two 1024-dimensional vectors, where For the first 512-dimensional vector, perform the following operations: 1. Convolution layer 1-1: convolution kernel size 3×3; number of convolution kernels 1024; stride: 1; 2. Convolutional layer 1-2: convolution kernel size 3×3; number of convolution kernels 1024; stride: 1; 3. Convolutional layers 1-3: kernel size 3×3; number of kernels 1024; Stride: 1; 4. Convolutional layers 1-4: kernel size 3×3; number of kernels 1024; Stride: 1; 5. Feature addition layer: Add the output vectors of convolutional layers 1-2 and 1-4 to obtain a 1024-dimensional vector; 6. Feature connection layer: connect the first 512-dimensional vector, the output of convolutional layers 1-2, and the data of the feature addition layer through channels, and output a 3072-dimensional vector; For the second 1024-dimensional vector, perform the following operations: 1. Perform channel connection with the output of the feature connection layer in step 6 above and output a 4096-dimensional vector; 2. Convolutional layer 2-1: convolution kernel size 3×3; number of convolution kernels 2048; stride: 1; Convolution upsampling 1: Use 2×2 transposed convolution upsampling, and the output channels are halved to 1024; Concatenation layer 1: connects the feature map with the output of activation layer 12, and the number of channels increases to 2048; Convolution layer 15: kernel size 3×3; number of kernels 1024; stride: 1; Activation layer 15: The activation function is LeakyReLU; Convolution layer 16: kernel size 3×3; number of kernels 1024; stride: 1; Activation layer 16: The activation function is LeakyReLU; Convolution upsampling 2: Use 2×2 transposed convolution upsampling, and the output channels are halved to 512; Concatenation layer 2: connects the feature map with the output of activation layer 10, and the number of channels increases to 1024; Convolution layer 17: kernel size 3×3; number of kernels 512; stride: 1; Activation layer 17: The activation function is LeakyReLU; Convolution layer 18: kernel size 3×3; number of kernels 512; stride: 1; Activation layer 18: The activation function is LeakyReLU; Convolution upsampling 3: Use 2×2 transposed convolution upsampling, and the output channels are halved to 256; Concatenation layer 3: connects the feature map with the output of activation layer 8, and the number of channels increases to 512; Convolution layer 19: kernel size 3×3; number of kernels 256; stride: 1; Activation layer 19: The activation function is LeakyReLU; Convolutional layer 20: convolution kernel size 3×3; number of convolution kernels 256; stride: 1; Activation layer 20: The activation function is LeakyReLU; Convolution upsampling 4: Use 2×2 transposed convolution upsampling, and the output channels are halved to 128; Concatenation layer 4: connects the feature map with the output of activation layer 6, and the number of channels increases to 256; Convolutional layer 21: convolution kernel size 3×3; number of convolution kernels 128; stride: 1; Activation layer 21: The activation function is LeakyReLU; Convolutional layer 22: convolution kernel size 3×3; number of convolution kernels 128; stride: 1; Activation layer 22: The activation function is LeakyReLU; Convolution upsampling 5: Use 2×2 transposed convolution upsampling, and the output channels are halved to 64; Concatenation layer 5: connects the feature map with the output of activation layer 4, and the number of channels increases to 128; Convolution layer 23: convolution kernel size 3×3; number of convolution kernels 64; stride: 1; Activation layer 23: The activation function is LeakyReLU; Convolutional layer 24: convolution kernel size 3×3; number of convolution kernels 64; stride: 1; Activation layer 24: the activation function is LeakyReLU; Convolution upsampling 6: Use 2×2 transposed convolution upsampling, and the output channels are halved to 32; Concatenation layer 6: connects the feature map with the output of activation layer 2, and the number of channels increases to 64; Convolutional layer 25: convolution kernel size 3×3; number of convolution kernels 32; stride: 1; Activation layer 25: The activation function is LeakyReLU; Convolution layer 26: convolution kernel size 3×3; number of convolution kernels 32; stride: 1; Activation layer 26: The activation function is LeakyReLU; Output layer: Type: Softmax; Number of neurons: 19; S4.4.2. Input the training set obtained in step S3 into the enhanced model D, calculate the weights and biases of the neural network units through the error back propagation algorithm, complete the training of the enhanced model D, and obtain the trained enhanced model D; The loss function of the enhanced model D is calculated as: ; in, is the loss function of the enhanced model D; is the total number of data samples in the reinforcement model D; For the id The true labels of samples, To strengthen the model D to predict the id The probability that a sample is class 1; S4.4.3. The node output values ​​of the 19 output units corresponding to the enhanced model D are d1, d2, d3, d4, ..., d19. The node output values ​​of the 19 output units are initially converted to obtain The corresponding probability value The calculation formula is: ; Then the node output values ​​of the 19 output units are further optimized and converted to obtain the output value The corresponding enhancement probability value The calculation formula is: ; in, G is the enhancement coefficient, and then the actual disease corresponding to each node is marked as ; S4.4.4. Adoption The training set obtained in step S3 is combined with the error chain derivation rule to train the enhanced model D. After the training is completed, the output values ​​of the 19 types of diseases corresponding to the enhanced model D are obtained as follows: .

[0021] S5. construct the loss function among the output layers of the basic model, the first auxiliary model, the second auxiliary model, and the enhanced model, and then construct the overall loss function among the basic model, the auxiliary model, and the enhanced model; Furthermore, the specific implementation method of step S5 includes the following steps: S5.1. Based on and , establish the loss function between the first auxiliary model B and the output layer of the basic model A, and the calculation formula is: ; based on and , construct the output value The corresponding probability value The loss function of the base model A , the calculation formula is: ; based on and , establish the relationship loss function between the output layer of the first auxiliary model B and the basic model A , the calculation formula is: ; in for The weight coefficient of for The weight coefficient of S5.2. Establish the relationship loss function between the output layer of the second auxiliary model C and the first auxiliary model B , the expression is: ; in, for The weight coefficient of for The weight coefficient of Based on and The loss function of Based on and The loss function of The calculation formula is: ; The calculation formula is: ; S5.3. Establish the relationship loss function between the output layer of the second auxiliary model C and the basic model A , the expression is: ; in, for The weight coefficient of for The weight coefficient of Based on and The loss function is Based on and The loss function of The calculation formula is: ; The calculation formula is: ; S5.4. Establishing the relationship loss function between the output layer of the enhanced model D and the base model A , the expression is: ; in, for The weight coefficient of for The weight coefficient of Based on and The loss function of The calculation formula is: ; S5.5. Establish the relationship loss function between the output layer of the reinforcement model D and the second auxiliary model C , the expression is: ; in, for The weight coefficient of for The weight coefficient of Based on and The loss function is Based on and The loss function of The calculation formula is: ; The calculation formula is: ; S5.6. Overall loss function among the constructed base model, first auxiliary model, second auxiliary model, and enhanced model , the calculation formula is: ;

[0022] S6. Using the data set for urban transportation infrastructure operation status detection and evaluation obtained in step S3, the constructed basic model, the first auxiliary model, the second auxiliary model, and the enhanced model are trained to obtain the trained basic model as the final intelligent identification model of transportation infrastructure operation status indicators;

[0023] Furthermore, the specific implementation method of step S6 is to use the data set for urban traffic infrastructure operation status detection and evaluation obtained in step S3, based on the chain derivation rule, and train the basic model, the first auxiliary model, the second auxiliary model, and the reinforcement model constructed for rutting recognition through error back propagation. When the overall loss function obtained in step S5 When is minimum, the corresponding model parameters are optimal, and the trained basic model A is the final intelligent identification model for transportation infrastructure operation status indicators.

[0024] S7. Constructing a digital twin model based on the intelligent identification model of the transportation infrastructure operation status indicator obtained in step S6; Furthermore, the specific implementation method of step S7 includes the following steps: S7.1. Export the intelligent identification model of traffic infrastructure operation status indicators obtained in step S6 as a deployment file for independent services, including one of a pt file, a pth file, a json file, and a txt file; S7.2. Build a transportation infrastructure information model; S7.2.1. Collect information on the location, length, width, and layout of transportation infrastructure; S7.2.2. The data obtained in step S7.2.1 is stored in a database, and a basic three-dimensional model of the transportation infrastructure is constructed, wherein the basic three-dimensional model of the transportation infrastructure includes geometric information of the transportation infrastructure; S7.2.3. Based on step S7.2.2, add material properties, design specifications, and historical maintenance data to the basic three-dimensional model of the transportation infrastructure to obtain a transportation infrastructure information model; S7.2.4. Export the transportation infrastructure information model to a format file supported by the web application, including a json file or a js file; S7.3. In the digital twin platform, import the deployment file of the intelligent identification model of the traffic infrastructure operation status indicator obtained in step S7.1, and import the traffic infrastructure information model format file obtained in step S7.2; S7.4. Use Python to construct an API interface between image data and the intelligent identification model of traffic infrastructure operation status indicators for intelligent identification of ruts; use Python to construct an API interface between the intelligent identification model of traffic infrastructure operation status indicators and the traffic infrastructure information model to complete the construction of the digital twin model.

[0025] S8. Based on the digital twin model obtained in step S7, a method for evaluating the operating status of transportation infrastructure is constructed to evaluate the operating status of transportation infrastructure.

[0026] Furthermore, the specific implementation method of step S8 includes the following steps: S8.1. Use a camera to collect road surface data and upload it to the digital twin platform. Then use the digital twin model to intelligently identify various indicators of transportation infrastructure in the image, select all images that include indicators of transportation infrastructure operation status, conduct on-site review of the indicators, and measure the indicators to obtain data that includes all indicators of transportation infrastructure operation status; S8.2. Propose a method for evaluating the operating status of facilities on roads; S8.2.1. Set based on crack length LCD , crack width LKD , rutting area CMJ , wave area BMJ , pit area KMJ , Oil spill area FMJ , Repair area XMJ , loose area SMJ , establish road operation status evaluation standards and divide the road operation status standards into 5 levels; S8.2.2. For each level of road operating condition standard in step S8.2.1, establish a road operating condition standard matrix. : , na The road operation status standard is graded as 1, 2, 3, 4, and 5. , , , , , , , They are the standard values ​​of the grades corresponding to crack length, crack width, rutting area, wave area, pothole area, oil spill area, repair area, and loose area; S8.2.3. Based on the collected images of the target road, a rapid intelligent identification method for the operating status indicators of urban transportation infrastructure is used to identify the yxCracks, ruts, waves, potholes, oil spills, repairs, loose diseases of the road, and measure and calculate the yx Length of cracks on roads , crack width , rutting area , wave area , pit area , Oil spill area , Repair area , loose area , and establish the road operation status matrix : ; S8.2.4. Comparison Matrix with 5 road operating conditions standards The standard value in is used to score each indicator value by expert scoring to construct a classification vector, where the expert scoring is divided into 1~5 points, including 1.5, 2.5, 3.5, and 4.5, where 1 means that the current indicator status is recommended to be divided into Class 1, 1.5 means that the current indicator status is recommended to be divided between Class 1 and Class 2, 2 means that the current indicator status is recommended to be divided into Class 2, 2.5 means that the current indicator status is recommended to be divided between Class 2 and Class 3, 3 means that the current indicator status is recommended to be divided into Class 3, 3.5 means that the current indicator status is recommended to be divided between Class 3 and Class 4, 4 means that the current indicator status is recommended to be divided into Class 4, 4.5 means that the current indicator status is recommended to be divided between Class 4 and Class 5, and 5 means that the current indicator status is recommended to be divided into Class 5; The road operation status scoring matrix is: ; in, ~ The scores represent the length of road cracks, the width of road cracks, the rutting area, the wave area, the pothole area, the oil spill area, the repair area, and the loose area; Based on the above scoring values, calculate ~ The weighted average score , the calculation formula is: ; in, Indicates the weight value corresponding to all road indicators; Compare The relationship between 1,2,3,4,5 and The closest value is the road operation status classification result, which is recorded as DL ; S8.3. Propose a method for evaluating the operating condition of facilities for bridges; S8.3.1. Set the length of cracks based on the bridge surface. QCD , crack width QKD , peeling area QBL , Steel bar corrosion area QXS , Seepage area QSS , Cell area QFW , pockmarked surface area QMM , demoulding defect area QTM , establish the bridge operation status evaluation standard and divide the bridge operation status standard into 5 levels; S8.3.2. For each level of bridge performance criteria, establish a bridge performance criteria matrix. , , nb The grade numbers for the bridge operation status standards are 1, 2, 3, 4, and 5; , , , , , , , They are the standard values ​​of the grades corresponding to the crack length, crack width, peeling area, steel bar corrosion area, water seepage area, honeycomb area, pitting area, and demoulding defect area; S8.3.3. Based on the target bridge image, use the rapid intelligent identification method of urban transportation infrastructure operation status indicators to identify the yx bridge cracks, spalling, steel bar corrosion, water seepage, honeycomb, pockmarks, demoulding defects, and measure and calculate the yx Length of cracks in bridges , crack width , peeling area , Steel bar corrosion area , Seepage area , Cell area , pockmarked surface area , demoulding defect area , and establish the yx Bridge operation status matrix : ; S8.3.4. Comparison Matrix with 5 bridge operating status standards The standard value in is used to score each indicator value by expert scoring to construct a classification vector, where the expert scoring is divided into 1~5 points, including 1.5, 2.5, 3.5, and 4.5, where 1 means that the current indicator status is recommended to be divided into Class 1, 1.5 means that the current indicator status is recommended to be divided between Class 1 and Class 2, 2 means that the current indicator status is recommended to be divided into Class 2, 2.5 means that the current indicator status is recommended to be divided between Class 2 and Class 3, 3 means that the current indicator status is recommended to be divided into Class 3, 3.5 means that the current indicator status is recommended to be divided between Class 3 and Class 4, 4 means that the current indicator status is recommended to be divided into Class 4, 4.5 means that the current indicator status is recommended to be divided between Class 4 and Class 5, and 5 means that the current indicator status is recommended to be divided into Class 5; The bridge operation status scoring matrix is: ; in, ~ They represent the scores of bridge crack length, bridge crack width, peeling area, steel bar corrosion area, water seepage area, honeycomb area, pitting area, and demoulding defect area; Based on the above scoring values, calculate ~ The weighted average score , the calculation formula is: ; in, Indicates the weight values ​​corresponding to all indicators of the bridge; comparison The relationship between 1,2,3,4,5 and The closest value is the classification result of the bridge operation status QL ; S8.4. Propose a method for evaluating the operating status of facilities in tunnels; S8.4.1. Set the crack length based on the tunnel surface SCD , crack width SKD , peeling area SBL , Steel bar corrosion area SXS , Seepage area SSS 、Number of drainage system blockages SSD , establish tunnel operation status evaluation standards and divide the tunnel operation status standards into 5 levels; S8.4.2. For each level of tunnel operating condition standard, establish a tunnel operating condition standard matrix. , , nc The level numbers for the tunnel operation status standards are 1, 2, 3, 4, and 5; , , , , , They are the standard values ​​corresponding to crack length, crack width, peeling area, steel bar corrosion area, water seepage area, and number of drainage system blockages; S8.4.3. Comparison Matrix with 5 tunnel operation status standards The standard value in is used to score each indicator value by expert scoring to construct a classification vector; the expert scoring is divided into 1~5 points, including 1.5, 2.5, 3.5, and 4.5, where 1 means that the current indicator status is recommended to be classified as category 1, 1.5 means that the current indicator status is recommended to be classified between category 1 and category 2, 2 means that the current indicator status is recommended to be classified as category 2, 2.5 means that the current indicator status is recommended to be classified between category 2 and category 3, 3 means that the current indicator status is recommended to be classified as category 3, 3.5 means that the current indicator status is recommended to be classified between category 3 and category 4, 4 means that the current indicator status is recommended to be classified as category 4, 4.5 means that the current indicator status is recommended to be classified between category 4 and category 5, and 5 means that the current indicator status is recommended to be classified as category 5; The tunnel operation status scoring matrix is: ; in, ~ The scores represent the length of tunnel cracks, the width of tunnel cracks, the spalling area, the steel bar corrosion area, the water seepage area, and the number of drainage system blockages; Based on the above scoring values, calculate ~ The weighted average score , the calculation formula is: ; in, Indicates the weight value corresponding to all tunnel indicators; Compare The relationship between 1,2,3,4,5 and The closest value is the tunnel operation status classification result SD ; S8.5. Based on the above calculation DL , QL , SD , calculate the comprehensive evaluation level of transportation infrastructure operation status , the calculation formula is: ; in, is the weight of the road operation status classification result, is the weight of the bridge operation status classification result, is the weight of the tunnel operation status classification result; At this point, the evaluation of the operating status of transportation infrastructure is completed.

[0027] The key points and protection points of the present invention are: (1) A method for rapid and intelligent identification of operating status indicators of urban transportation infrastructure (2) A method for evaluating the operating status of transportation infrastructure It should be noted that relational terms such as "first" and "second" are used only to distinguish one entity or operation from another entity or operation, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Moreover, the terms "include", "comprise" or any other variants thereof are intended to cover non-exclusive inclusion, so that a process, method, article or device including a series of elements includes not only those elements, but also other elements not explicitly listed, or also includes elements inherent to such process, method, article or device. In the absence of further restrictions, the elements defined by the sentence "comprise a ..." do not exclude the existence of other identical elements in the process, method, article or device including the elements.

[0028] Although the present application has been described above with reference to specific embodiments, various modifications may be made thereto and parts thereof may be replaced with equivalents without departing from the scope of the present application. In particular, as long as there is no structural conflict, the various features in the specific embodiments disclosed in the present application may be used in combination with each other in any manner, and the fact that these combinations are not exhaustively described in this specification is only for the sake of omitting space and saving resources. Therefore, the present application is not limited to the specific embodiments disclosed herein, but includes all technical solutions falling within the scope of the claims.

Claims

1. A method for detecting and evaluating the operating status of urban transportation infrastructure, characterized in that: The steps include: S1. Using a vehicle-mounted camera, collect the surface images of traffic infrastructure, including the surface images of roads, bridges, and tunnels, while the vehicle is driving; S2. Set indicators for the operation status of transportation infrastructure; S3. For the surface images of the transportation infrastructure collected in step S1, select all images including the operating status indicators of the transportation infrastructure, annotate the operating status indicators by manual annotation, and then construct a data set for detecting and evaluating the operating status of urban transportation infrastructure; S4. Construct an intelligent identification model for traffic infrastructure operation status indicators, including a basic model, a first auxiliary model, a second auxiliary model, and an enhanced model; S5. construct the loss function among the output layers of the basic model, the first auxiliary model, the second auxiliary model, and the enhanced model, and then construct the overall loss function among the basic model, the auxiliary model, and the enhanced model; S6. Using the data set for urban transportation infrastructure operation status detection and evaluation obtained in step S3, the constructed basic model, the first auxiliary model, the second auxiliary model, and the enhanced model are trained to obtain the trained basic model as the final intelligent identification model of transportation infrastructure operation status indicators; S7. Constructing a digital twin model based on the intelligent identification model of the transportation infrastructure operation status indicator obtained in step S6; S8. Based on the digital twin model obtained in step S7, a method for evaluating the operating status of transportation infrastructure is constructed to evaluate the operating status of transportation infrastructure.

2. A method for detecting and evaluating the operating status of urban transportation infrastructure, characterized in that: Among the traffic infrastructure operation status indicators in step S2, the operation status indicators of roads include: cracks, rutting, waves, potholes, oil spills, repairs, and looseness; the operation status indicators of bridges include: cracks, spalling, steel corrosion, water seepage, honeycombs, rough surfaces, and demoulding defects; the operation status indicators of tunnels include: cracks, spalling, steel corrosion, water seepage, and drainage system blockage.

3. A method for detecting and evaluating the operating status of urban transportation infrastructure according to claim 2, characterized in that: The specific implementation method of step S3 includes the following steps: S3.

1. Manually select the surface images of the transportation infrastructure collected in step S1, and then manually label the operating status indicators of roads, bridges, and tunnels; S3.

2. The labeled traffic infrastructure appearance images are used as a dataset for urban traffic infrastructure operation status detection and evaluation, and are divided into a training set and a test set in a ratio of 7:

3.

4. A method for detecting and evaluating the operating status of urban transportation infrastructure according to claim 3, characterized in that: The specific implementation method of step S4 includes the following steps: S4.

1. Basic model for constructing intelligent identification model of traffic infrastructure operation status indicators: S4.1.

1. The basic model A consists of 1 input layer, 14 convolutional layers, 14 activation layers, 5 pooling layers, 1 feature splitting layer, 4 concatenation layers, 4 convolutional upsampling layers, and 1 output layer. S4.1.

2. Input the training set obtained in step S3 into the basic model A, calculate the weights and biases of the neural network units through the error back propagation algorithm, complete the training of the basic model A, and obtain the trained basic model A; The calculation formula of the loss function of the basic model A is: ； in, is the loss function of the basic model A; is the total number of data samples in the basic model A; For the ia The true labels of samples, Based on the prediction of model A ia The probability that a sample is class 1; S4.1.

3. The node output values ​​of the 19 output units corresponding to the basic model A are a1, a2, a3, a4, ..., a19. The node output values ​​of the 19 output units are initially converted to obtain the output values The corresponding probability value The calculation formula is: ； In order to amplify the probability difference between the output values ​​and enhance the model's ability to distinguish, the 19 node outputs are further optimized and converted into: ； in, G is the enhancement coefficient, and then the actual disease corresponding to each node is marked as ; S4.1.

4. Adoption The training set obtained in step S3 is combined with the error chain derivation rule to train the basic model A. After the training is completed, the output values ​​of the 19 types of diseases corresponding to the basic model A are obtained, which are recorded as , , , , …, ; S4.

2. Constructing the first auxiliary model of the intelligent identification model of traffic infrastructure operation status indicators: S4.2.

1. The first auxiliary model B consists of 1 input layer, 18 convolutional layers, 18 activation layers, 5 pooling layers, 1 feature splitting layer, 4 concatenation layers, 4 convolutional upsampling layers, and 1 output layer; S4.2.

2. Input the training set obtained in step S3 into the first auxiliary model B, calculate the weights and biases of the neural network units through the error back propagation algorithm, complete the training of the first auxiliary model B, and obtain the trained first auxiliary model B; The calculation formula of the loss function of the first auxiliary model B is: ； in, is the loss function of the first auxiliary model B; is the total number of data samples in the first auxiliary model B; For the ib The true labels of samples, The first auxiliary model B predicts ib The probability that a sample is class 1; S4.2.

3. The node output values ​​of the 19 output units corresponding to the first auxiliary model B are b1, b2, b3, b4, ..., b19. The node output values ​​of the 19 output units are initially converted to obtain the output values b i The corresponding probability value The calculation formula is: ； Then the node output values ​​of the 19 output units are further optimized and converted to obtain the output value b i The corresponding enhancement probability value The calculation formula is: ； S4.2.

4. Adoption The training set obtained in step S3 is combined with the error chain derivation rule to train the first auxiliary model B. After the training is completed, the disease output value corresponding to the first auxiliary model B is obtained as , among the 19 nodes in the output layer, the actual disease corresponding to each node is marked as , i =1,2,3,…,19; S4.

3. Constructing the second auxiliary model of the intelligent identification model of traffic infrastructure operation status indicators: S4.3.

1. The second auxiliary model C consists of 1 input layer, 22 convolutional layers, 22 activation layers, 6 pooling layers, 1 feature splitting layer, 5 concatenation layers, 5 convolution upsampling layers, and 1 output layer; S4.3.

2. Input the training set obtained in step S3 into the second auxiliary model C, calculate the weights and biases of the neural network units through the error back propagation algorithm, complete the training of the second auxiliary model C, and obtain the trained second auxiliary model C; The calculation formula of the loss function of the second auxiliary model C is: ； in, is the loss function of the second auxiliary model C; is the total number of data samples in the second auxiliary model C; For the ic The true labels of samples, The second auxiliary model C predicts ic The probability that a sample is class 1; S4.3.

3. The node output values ​​of the 19 output units corresponding to the second auxiliary model C are c1, c2, c3, c4, ..., c19. The node output values ​​of the 19 output units are initially converted to obtain The corresponding probability value The calculation formula is: ； Then the node output values ​​of the 19 output units are further optimized and converted to obtain the output value The corresponding enhancement probability value The calculation formula is: ； in, G is the enhancement coefficient, and then the actual disease corresponding to each node is marked as ; S4.3.

4. Adoption The training set obtained in step S3 is combined with the error chain derivation rule to train the second auxiliary model C. After the training is completed, the output values ​​of the 19 types of diseases corresponding to the second auxiliary model C are obtained as follows: ; S4.

4. Constructing an enhanced model for intelligent identification of traffic infrastructure operation status indicators: S4.4.

1. The enhanced model D consists of 1 input layer, 26 convolutional layers, 26 activation layers, 7 pooling layers, 1 feature splitting layer, 6 concatenation layers, 6 convolutional upsampling layers, and 1 output layer. S4.4.

2. Input the training set obtained in step S3 into the enhanced model D, calculate the weights and biases of the neural network units through the error back propagation algorithm, complete the training of the enhanced model D, and obtain the trained enhanced model D; The loss function of the enhanced model D is calculated as: ； in, is the loss function of the enhanced model D; is the total number of data samples in the reinforcement model D; For the id The true labels of samples, To strengthen the model D to predict the id The probability that a sample is class 1; S4.4.

3. The node output values ​​of the 19 output units corresponding to the enhanced model D are d1, d2, d3, d4, ..., d19. The node output values ​​of the 19 output units are initially converted to obtain The corresponding probability value The calculation formula is: ； Then the node output values ​​of the 19 output units are further optimized and converted to obtain the output value The corresponding enhancement probability value The calculation formula is: ； in, G is the enhancement coefficient, and then the actual disease corresponding to each node is marked as ; S4.4.

4. Adoption The training set obtained in step S3 is combined with the error chain derivation rule to train the enhanced model D. After the training is completed, the output values ​​of the 19 types of diseases corresponding to the enhanced model D are obtained as follows: .

5. A method for detecting and evaluating the operating status of urban transportation infrastructure according to claim 4, characterized in that: The specific implementation method of step S5 includes the following steps: S5.

1. Based on and , establish the loss function between the first auxiliary model B and the output layer of the basic model A, and the calculation formula is: ； based on and , construct the output value The corresponding probability value The loss function of the base model A , the calculation formula is: ； based on and , establish the relationship loss function between the output layer of the first auxiliary model B and the basic model A , the calculation formula is: ； in for The weight coefficient of for The weight coefficient of S5.

2. Establish the relationship loss function between the output layer of the second auxiliary model C and the first auxiliary model B , the expression is: ； in, for The weight coefficient of for The weight coefficient of Based on and The loss function of Based on and The loss function of The calculation formula is: ； The calculation formula is: ； S5.

3. Establish the relationship loss function between the output layer of the second auxiliary model C and the basic model A , the expression is: ； in, for The weight coefficient of for The weight coefficient of Based on and The loss function is Based on and The loss function of The calculation formula is: ； The calculation formula is: ； S5.

4. Establishing the relationship loss function between the output layer of the enhanced model D and the base model A , the expression is: ； in, for The weight coefficient of for The weight coefficient of Based on and The loss function of The calculation formula is: ； S5.

5. Establish the relationship loss function between the output layer of the reinforcement model D and the second auxiliary model C , the expression is: ； in, for The weight coefficient of for The weight coefficient of Based on and The loss function is Based on and The loss function of The calculation formula is: ； The calculation formula is: ； S5.

6. Overall loss function between the constructed base model, first auxiliary model, second auxiliary model, and enhanced model , the calculation formula is: 。 6. A method for detecting and evaluating the operating status of urban transportation infrastructure according to claim 5, characterized in that: The specific implementation method of step S6 is to use the data set for urban traffic infrastructure operation status detection and evaluation obtained in step S3, and train the basic model, the first auxiliary model, the second auxiliary model, and the reinforcement model constructed for rutting recognition through error back propagation based on the chain derivative rule. When the overall loss function obtained in step S5 When is minimum, the corresponding model parameters are optimal, and the trained basic model A is the final intelligent identification model for transportation infrastructure operation status indicators.

7. A method for detecting and evaluating the operating status of urban transportation infrastructure according to claim 6, characterized in that: The specific implementation method of step S7 includes the following steps: S7.

1. Export the intelligent identification model of traffic infrastructure operation status indicators obtained in step S6 as a deployment file for independent services, including one of a pt file, a pth file, a json file, and a txt file; S7.

2. Build a transportation infrastructure information model; S7.2.

1. Collect information on the location, length, width, and layout of transportation infrastructure; S7.2.

2. The data obtained in step S7.2.1 is stored in a database, and a basic three-dimensional model of the transportation infrastructure is constructed, wherein the basic three-dimensional model of the transportation infrastructure includes geometric information of the transportation infrastructure; S7.2.

3. Based on step S7.2.2, add material properties, design specifications, and historical maintenance data to the basic three-dimensional model of the transportation infrastructure to obtain a transportation infrastructure information model; S7.2.

4. Export the transportation infrastructure information model to a format file supported by the web application, including a json file or a js file; S7.

3. In the digital twin platform, import the deployment file of the intelligent identification model of the traffic infrastructure operation status indicator obtained in step S7.1, and import the traffic infrastructure information model format file obtained in step S7.2; S7.

4. Use Python to construct an API interface between image data and the intelligent identification model of traffic infrastructure operation status indicators for intelligent identification of ruts; use Python to construct an API interface between the intelligent identification model of traffic infrastructure operation status indicators and the traffic infrastructure information model to complete the construction of the digital twin model.

8. A method for detecting and evaluating the operating status of urban transportation infrastructure according to claim 7, characterized in that: The specific implementation method of step S8 includes the following steps: S8.

1. Use a camera to collect road surface data and upload it to the digital twin platform. Then use the digital twin model to intelligently identify various indicators of transportation infrastructure in the image, select all images that include indicators of transportation infrastructure operation status, conduct on-site review of the indicators, and measure the indicators to obtain data that includes all indicators of transportation infrastructure operation status; S8.

2. Propose a method for evaluating the operating status of facilities on roads; S8.2.

1. Set based on crack length LCD , crack width LKD , rutting area CMJ , wave area BMJ , pit area KMJ , Oil spill area FMJ , Repair area XJ , loose area SMJ , establish road operation status evaluation standards and divide the road operation status standards into 5 levels; S8.2.

2. For each level of road operating condition standard in step S8.2.1, establish a road operating condition standard matrix. : , na The road operation status standard is graded as 1, 2, 3, 4, and 5. , , , , , , , They are the standard values ​​of the grades corresponding to crack length, crack width, rutting area, wave area, pothole area, oil spill area, repair area, and loose area; S8.2.

3. Based on the collected images of the target road, a rapid intelligent identification method for the operating status indicators of urban transportation infrastructure is used to identify the yx Cracks, ruts, waves, potholes, oil spills, repairs, loose diseases of the road, and measure and calculate the yx Length of cracks on roads , crack width , rutting area , wave area , pit area , Oil spill area , Repair area , loose area , and establish the road operation status matrix : ； S8.2.

4. Comparison Matrix with 5 road operating conditions standards The standard value in is used to score each indicator value by expert scoring to construct a classification vector, where the expert scoring is divided into 1~5 points, including 1.5, 2.5, 3.5, and 4.5, where 1 means that the current indicator status is recommended to be divided into Class 1, 1.5 means that the current indicator status is recommended to be divided between Class 1 and Class 2, 2 means that the current indicator status is recommended to be divided into Class 2, 2.5 means that the current indicator status is recommended to be divided between Class 2 and Class 3, 3 means that the current indicator status is recommended to be divided into Class 3, 3.5 means that the current indicator status is recommended to be divided between Class 3 and Class 4, 4 means that the current indicator status is recommended to be divided into Class 4, 4.5 means that the current indicator status is recommended to be divided between Class 4 and Class 5, and 5 means that the current indicator status is recommended to be divided into Class 5; The road operation status scoring matrix is: ; in, ~ The scores represent the length of road cracks, the width of road cracks, the rutting area, the wave area, the pothole area, the oil spill area, the repair area, and the loose area; Based on the above scoring values, calculate ~ The weighted average score , the calculation formula is: ； in, Indicates the weight value corresponding to all road indicators; Compare The relationship between 1,2,3,4,5 and The closest value is the road operation status classification result, which is recorded as DL ; S8.

3. Propose a method for evaluating the operating condition of facilities for bridges; S8.3.

1. Set the length of cracks based on the bridge surface. QCD , crack width QKD , peeling area QBL , Steel bar corrosion area QXS , Seepage area QSS , Cell area QW , pockmarked surface area QMM , demoulding defect area QTM , establish the bridge operation status evaluation standard and divide the bridge operation status standard into 5 levels; S8.3.

2. For each level of bridge performance criteria, establish a bridge performance criteria matrix. , , nb The grade numbers for the bridge operation status standards are 1, 2, 3, 4, and 5; , , , , , , , They are the standard values ​​of the grades corresponding to the crack length, crack width, peeling area, steel bar corrosion area, water seepage area, honeycomb area, pitting area, and demoulding defect area; S8.3.

3. Based on the collected images of the target bridge, a rapid intelligent identification method for the operating status indicators of urban transportation infrastructure is used to identify the yx bridge cracks, spalling, steel bar corrosion, water seepage, honeycomb, pockmarks, demoulding defects, and measure and calculate the yx Length of cracks in bridges , crack width , peeling area , Steel bar corrosion area , Seepage area , Cell area , pockmarked surface area , demoulding defect area , and establish the yx Bridge operation status matrix : ； S8.3.

4. Comparison Matrix with 5 bridge operating status standards The standard value in is used to score each indicator value by expert scoring to construct a classification vector, where the expert scoring is divided into 1~5 points, including 1.5, 2.5, 3.5, and 4.5, where 1 means that the current indicator status is recommended to be divided into Class 1, 1.5 means that the current indicator status is recommended to be divided between Class 1 and Class 2, 2 means that the current indicator status is recommended to be divided into Class 2, 2.5 means that the current indicator status is recommended to be divided between Class 2 and Class 3, 3 means that the current indicator status is recommended to be divided into Class 3, 3.5 means that the current indicator status is recommended to be divided between Class 3 and Class 4, 4 means that the current indicator status is recommended to be divided into Class 4, 4.5 means that the current indicator status is recommended to be divided between Class 4 and Class 5, and 5 means that the current indicator status is recommended to be divided into Class 5; The bridge operation status scoring matrix is: ; in, ~ They represent the scores of bridge crack length, bridge crack width, peeling area, steel bar corrosion area, water seepage area, honeycomb area, pitting area, and demoulding defect area; Based on the above scoring values, calculate ~ The weighted average score , the calculation formula is: ； in, Indicates the weight values ​​corresponding to all indicators of the bridge; comparison The relationship between 1,2,3,4,5 and The closest value is the classification result of the bridge operation status QL ; S8.

4. Propose a method for evaluating the operating status of facilities in tunnels; S8.4.

1. Set the crack length based on the tunnel surface SCD , crack width SKD , peeling area SBL , Steel bar corrosion area SXS , Seepage area SSS 、Number of drainage system blockages SSD , establish tunnel operation status evaluation standards and divide the tunnel operation status standards into 5 levels; S8.4.

2. For each level of tunnel operating condition standard, establish a tunnel operating condition standard matrix. , , nc The level numbers for the tunnel operation status standards are 1, 2, 3, 4, and 5; , , , , , They are the standard values ​​corresponding to crack length, crack width, peeling area, steel bar corrosion area, water seepage area, and number of drainage system blockages; S8.4.

3. Comparison Matrix with 5 tunnel operation status standards The standard value in is used to score each indicator value by expert scoring to construct a classification vector; the expert scoring is divided into 1~5 points, including 1.5, 2.5, 3.5, and 4.5, where 1 means that the current indicator status is recommended to be classified as category 1, 1.5 means that the current indicator status is recommended to be classified between category 1 and category 2, 2 means that the current indicator status is recommended to be classified as category 2, 2.5 means that the current indicator status is recommended to be classified between category 2 and category 3, 3 means that the current indicator status is recommended to be classified as category 3, 3.5 means that the current indicator status is recommended to be classified between category 3 and category 4, 4 means that the current indicator status is recommended to be classified as category 4, 4.5 means that the current indicator status is recommended to be classified between category 4 and category 5, and 5 means that the current indicator status is recommended to be classified as category 5; The tunnel operation status scoring matrix is: ; in, ~ The scores represent the length of tunnel cracks, the width of tunnel cracks, the spalling area, the steel bar corrosion area, the water seepage area, and the number of drainage system blockages; Based on the above scoring values, calculate ~ The weighted average score , the calculation formula is: ； in, Indicates the weight value corresponding to all tunnel indicators; Compare The relationship between 1,2,3,4,5 and The closest value is the tunnel operation status classification result SD ; S8.

5. Based on the above calculation DL , QL , SD , calculate the comprehensive evaluation level of transportation infrastructure operation status , the calculation formula is: ； in, is the weight of the road operation status classification result, is the weight of the bridge operation status classification result, is the weight of the tunnel operation status classification result; At this point, the evaluation of the operating status of transportation infrastructure is completed.

Images