A method and system for detecting and estimating target bridge lane load based on remote sensing images

Abstract

The application relates to a method and system for detecting and estimating target bridge lane load based on remote sensing images, which comprises the following steps: collecting remote sensing image sets of a region to be studied to extract bridge and standard load vehicle characteristic information; based on the obtained bridge and vehicle characteristic information, obtaining a bridge training sample set and a standard load vehicle image training sample set of the region to be studied according to artificial visual interpretation; training a deep learning model based on the training sample set to construct an identification model; preliminarily identifying and matching the corresponding levels of beam bridges in the region to be studied and standard load vehicles in the region to be studied based on the identification model; obtaining the remote sensing images of the target bridge according to the recorded spatial position information based on the target bridge obtained through preliminary matching to extract and calculate the span, and obtaining the basic information of the target bridge; and estimating the target bridge lane load based on the basic information of the target bridge. The application solves the problem of estimating the bridge vehicle load based on remote sensing images.

Description

Technical Field

[0001] This application relates to the field of remote sensing detection technology, and in particular to a method and system for estimating the lane load of a target bridge based on remote sensing image detection. Background Technology

[0002] Remote sensing, as a comprehensive Earth observation technology, possesses unparalleled advantages over other technologies, including wide coverage, large information volume, and high timeliness. Currently, remote sensing technology is widely used in various fields such as terrain monitoring, water and soil resource surveys, and road monitoring in uninhabited or data-scarce areas. However, in research on acquiring information from areas lacking data, scholars often rely on geospatial and spectral features to obtain parameters and apply them to model or mechanism studies. This research approach is complex and lacks timeliness, and few studies apply remote sensing to estimating attribute information such as urban road infrastructure. Furthermore, in the era of remote sensing big data, deep learning methods, due to their strong learning capabilities, ability to acquire feature information of ground objects for object identification, wide coverage, and capacity for large-scale data processing and computation via computers, are widely used in fields such as ground object recognition, scene recognition, and facial recognition.

[0003] While deep learning can acquire information from areas lacking data based on big data, its application in urban bridge load estimation still faces some challenges. Firstly, deep learning requires a large number of samples for model training to extract target features, but current technology cannot support obtaining attribute information of target features, such as lane loads on urban roads and bridges. Secondly, deep learning combined with remote sensing imagery is easily limited by the quality of the imagery, often missing effective features of ground objects, leading to inaccurate information acquisition and hindering precise estimation of urban bridge lane loads, thus reducing work efficiency. Summary of the Invention

[0004] This application provides a method and system for estimating the lane load of a target bridge based on remote sensing imagery. Utilizing existing remote sensing imagery data and actual data, a deep learning model is constructed with urban bridges as the primary identification target. This model enables the classification and monitoring of urban bridges, and the identification and extraction of urban bridges in specific areas. It addresses the lack of a database of urban bridges and their standard heavy-duty vehicles. Finally, by combining information such as the net span and number of lanes of urban bridges extracted from UAV remote sensing images, the system effectively estimates the lane load capacity of urban bridges. This solves the problem of estimating bridge vehicle loads based on insufficient data from remote sensing images, providing technical support for obtaining road traffic information and route planning in special and complex geographical environments, and improving work efficiency.

[0005] In a first aspect, this application provides a method for estimating the lane load of a target bridge based on remote sensing image detection, comprising the following steps: acquiring a set of remote sensing images of the area to be studied to extract feature information of the bridge and its standard load vehicle; based on the obtained feature information of the bridge and standard load vehicle, obtaining a training sample set of bridge images and a training sample set of standard load vehicle images of the area to be studied by manual visual interpretation; training a deep learning model based on the training sample set to construct a recognition model; performing preliminary identification and matching of beam bridges and standard load vehicle corresponding levels in the area to be studied based on the recognition model to determine the specific level information of beam bridges in the area to be studied and recording their image spatial location information; obtaining remote sensing images of the target bridge based on the preliminary matching using a UAV according to the recorded image spatial location information to obtain the basic information of the target bridge, wherein the basic information is used to display the basic construction data of the target bridge; and estimating the lane load of the target bridge based on the basic information of the target bridge according to bridge design standards.

[0006] Optionally, a remote sensing image set of the area to be studied is collected to extract the feature information of bridges and their standard load vehicles, including: collecting the actual bridge construction environment of the area to be studied and obtaining the corresponding remote sensing images; performing data preprocessing on the remote sensing images, and cropping them according to visual interpretation to obtain the remote sensing image set of the area to be studied, so as to extract the feature information of bridges and their standard load vehicles in the area to be studied.

[0007] Optionally, based on the obtained bridge and standard load vehicle feature information, a bridge training sample set and a standard load vehicle image training sample set for the study area are obtained through manual visual interpretation. This includes: obtaining typical bridge images containing lane number information within the study area based on bridge feature information through manual visual interpretation as the bridge training sample set; obtaining standard load vehicle images of different levels within the study area through manual visual interpretation based on standard load vehicle feature information, combined with prior knowledge and vehicle management data, as standard load vehicle image training sample sets, wherein the standard load vehicles include Class A standard load vehicles and Class B standard load vehicles.

[0008] Optionally, training the deep learning model based on the training sample set to construct a recognition model includes: using the training sample set as input data based on the U-Net model to train and optimize the model in order to construct a recognition model.

[0009] Optionally, based on the recognition model, preliminary identification and matching of beam bridges and standard heavy-duty trucks in the study area are performed to determine the specific grade information of beam bridges in the study area and record their image spatial location information. This includes: determining whether there is a beam bridge in the image under study; if so, verifying the time attribute of the identified beam bridge and determining the bridge construction time range to match the calculation formula in the bridge design standard determined within the corresponding time period; otherwise, discarding the image; based on the determination that there is a beam bridge, further determining whether there are Class A standard heavy-duty trucks and / or Class B standard heavy-duty trucks in the image under study; if not, discarding the image; if so, if a beam bridge and a Class A standard heavy-duty truck appear simultaneously in the image, the bridge is determined to be a Class A bridge; if a beam bridge, a Class A standard heavy-duty truck, and a Class B standard heavy-duty truck appear simultaneously in the image, the bridge is determined to be a Class A bridge; if a beam bridge and a Class B standard heavy-duty truck appear simultaneously in the image, the bridge is determined to be a Class B bridge; and recording the spatial location information of the beam bridges whose time attributes have been determined and verified.

[0010] Optionally, it can be determined whether there is a beam bridge in the image under study. If so, the time attribute of the identified beam bridge is verified to determine the bridge construction time range, so as to match the calculation formula in the bridge design standard determined within the corresponding time period. This includes: based on the image under study, first confirming whether there is a beam bridge. If not, continue to select images under study according to specific time intervals as needed to confirm whether there is a beam bridge. If there is, obtain the image time of the image under study. Based on the obtained image time, determine the time range of the beam bridge during design or construction, so as to match the calculation formula in the bridge design standard determined within the corresponding time period. The bridge design standard is continuously updated and optimized over time, and each set of bridge design standards has different provisions for standard heavy-duty trucks.

[0011] Optionally, based on the target bridge obtained through preliminary matching, a remote sensing image of the target bridge is obtained by an unmanned aerial vehicle (UAV) according to the recorded spatial location information of the image, so as to obtain the basic information of the target bridge. This includes: acquiring a two-dimensional side view image of the target bridge in the area to be studied and performing image processing; calculating the net span of the target bridge based on the scaling ratio of the UAV remote sensing image and the distance in the image; and obtaining the number of lanes and piers through manual visual interpretation to obtain the basic information of the target bridge. The basic information includes the bridge classification, net span, number of piers, and number of lanes in the area to be studied.

[0012] Optionally, the lane load of the target bridge is estimated based on the basic information of the target bridge according to the bridge design standard, including: substituting the obtained basic information of the target bridge into the bridge lane load calculation formula in the corresponding bridge design standard to estimate the lane load of the target bridge, wherein the calculation formula is expressed as: total vehicle load of the target bridge = number of lanes × single lane load × load factor × number of spans.

[0013] Secondly, this application provides a system for estimating the lane load of a target bridge based on remote sensing image detection. The system includes: an acquisition module for collecting a set of remote sensing images of the area under study to extract feature information of the bridge and its standard-load vehicle; a training sample module for obtaining a bridge training sample set and a standard-load vehicle image training sample set of the area under study based on the obtained feature information of the bridge and standard-load vehicle, according to manual visual interpretation; a model construction module for training the deep learning model based on the training sample set to construct a recognition model; a matching module for performing preliminary identification and matching between beam bridges and standard-load vehicle levels in the area under study based on the recognition model, to determine the specific level information of beam bridges in the area under study and record their image spatial location information; a span extraction module for obtaining remote sensing images of the target bridge using a UAV based on the recorded image spatial location information, based on the preliminarily matched target bridge, to obtain the basic information of the target bridge, wherein the basic information is used to display the basic construction data of the target bridge; and a load estimation module for estimating the lane load of the target bridge based on the basic information of the target bridge according to bridge design standards.

[0014] Thirdly, this application also provides a computer device, including a memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that the processor executes the computer program to implement the steps of the method described above.

[0015] Fourthly, a computer-readable storage medium is provided, on which a computer program is stored, which, when executed by a processor, implements the steps of the method described above.

[0016] This application has at least the following advantages:

[0017] By collecting remote sensing image sets and actual data of the area under study, characteristic information of bridges and standard-load vehicles in the area was obtained, resulting in a training sample set. A recognition model was constructed using deep learning algorithms to identify and match beam bridges with vehicles, determine the grade of beam bridges, and realize the identification and extraction of bridges in specific areas. This solved the problem of missing databases of urban bridges and their standard-load vehicles. Furthermore, UAV remote sensing imagery was used to supplement and verify satellite remote sensing imagery to obtain basic information such as the span, length, and width of target bridges. Based on the bridge design standards corresponding to the target bridges, the lane load capacity of bridges in the area under study was effectively estimated. This provides technical support for obtaining road traffic information and route planning in special and complex geographical environments, and improves work efficiency. Attached Figure Description

[0018] Figure 1 This is a diagram illustrating the application environment of a method for estimating the lane load of a target bridge based on remote sensing imagery in one embodiment.

[0019] Figure 2 This is a flowchart illustrating the steps of a method for estimating the lane load of a target bridge based on remote sensing imagery in one embodiment.

[0020] Figure 3 Here is a flowchart illustrating a method for estimating the lane load of a target bridge based on remote sensing imagery in one embodiment;

[0021] Figure 4 This is a flowchart illustrating the process of extracting feature information of a bridge and its standard load vehicles in one embodiment.

[0022] Figure 5 Here is a flowchart illustrating the process of constructing a training sample set in one embodiment;

[0023] Figure 6 This is a schematic diagram of a multi-span beam bridge shown in one embodiment;

[0024] Figure 7 This embodiment shows an example diagram of a heavy-duty truck conforming to bridge design standards;

[0025] Figure 8 This is a schematic diagram illustrating the identification and matching of bridges and standard heavy-duty vehicles in the area under study in one embodiment;

[0026] Figure 9 This is a schematic diagram of lane load in one embodiment;

[0027] Figure 10 Here is a structural block diagram of a system for estimating the lane load of a target bridge based on remote sensing imagery, as shown in one embodiment.

[0028] Figure 11 This is a schematic structural diagram of a computer device in one embodiment. Detailed Implementation

[0029] The present application will be further described in detail below with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely illustrative of the present application and are not intended to limit the scope of the present application.

[0030] It should be noted that the terminology used herein is for the purpose of describing particular implementations only and is not intended to limit the exemplary implementations according to this application. As used herein, the singular form is intended to include the plural form as well, unless the context clearly indicates otherwise. Furthermore, it should be understood that when used in this specification, the words “comprising” and / or “including” indicate the presence of features, steps, operations, devices, components, and / or combinations thereof.

[0031] To make the objectives, technical solutions, and advantages of the embodiments of this application clearer, the various embodiments of this application will be described in detail below with reference to the accompanying drawings. However, those skilled in the art will understand that many technical details have been provided in the embodiments of this application to facilitate a better understanding of the application. However, the technical solutions claimed in this application can be implemented even without these technical details and various variations and modifications based on the following embodiments. The reference numerals in the following embodiments are for descriptive convenience and should not constitute any limitation on the specific implementation of this application. The various embodiments can be combined with and referenced by each other without contradiction.

[0032] Figure 2 This application provides a flowchart illustrating a method for estimating the lane load of a target bridge based on remote sensing image detection, which can be implemented by, for example... Figure 1 The user-space file server in the system shown is executed. For example... Figure 2 , Figure 3 As shown, the method may include the following steps:

[0033] S201. Collect remote sensing image sets of the area to be studied to extract characteristic information of bridges and their standard load vehicles.

[0034] S202. Based on the obtained bridge and standard load vehicle feature information, the bridge training sample set and standard load vehicle image training sample set of the area to be studied are obtained by manual visual interpretation.

[0035] S203. Train the deep learning model based on the training sample set to build a recognition model;

[0036] S204. Based on the recognition model, perform preliminary identification and matching of beam bridges and standard heavy-duty trucks in the study area to determine the specific grade information of beam bridges in the study area and record their image spatial location information.

[0037] S205. Based on the target bridge obtained through preliminary matching, remote sensing images of the target bridge are obtained by UAV according to the recorded image spatial location information, so as to obtain the basic information of the target bridge. The basic information is used to display the basic construction data of the target bridge.

[0038] S206. Estimate the lane load of the target bridge based on the basic information of the target bridge and the bridge design standards.

[0039] This application primarily addresses existing remote sensing imagery and real-world data by providing a method for estimating urban bridge loads using deep learning and target detection based on geographic thinking. This method enables the classification and monitoring of urban bridges. Finally, by combining information such as the net span and number of lanes of urban bridges extracted from UAV remote sensing imagery, it effectively estimates the lane load capacity of urban bridges. Based on simulation and verification within a study area at the urban scale, this method is applied to bridge identification in special scenarios, providing technical support for collecting information on environmental elements in special scenarios.

[0040] The following is a detailed explanation of each step:

[0041] Please refer to Figure 2 , Figure 3 As shown, step S201 involves acquiring a set of remote sensing images of the area to be studied to extract the characteristic information of the bridge and its standard load vehicle.

[0042] In this embodiment, it should be noted that the actual bridge construction environment in the area to be studied is collected, and corresponding remote sensing images are obtained; the remote sensing images are preprocessed, and a sample remote sensing image set of the study area is obtained by cropping according to visual interpretation, which includes large parking lots and bridge target features in the area to be studied, so as to extract the feature information of bridges and their standard load vehicles in the area to be studied.

[0043] Specifically, refer to Figure 4 As shown, the collected satellite remote sensing data is Sentinel-1 series data. After image preprocessing such as atmospheric correction, noise removal, impact enhancement, and fusion, a remote sensing image set of the study area was obtained based on visual interpretation and cropped, serving as the foundation data for constructing various sample datasets. Manual interpretation is currently the most widely used image extraction method in China, such as in land surveys, geological surveys, and water conservancy surveys. Specifically, it refers to the use of differences in brightness or pixel values ​​and spatial variations in remote sensing images to represent the differences between different land features, which is the physical basis for distinguishing different image features. The remote sensing image set here should spatially include target features such as large parking lots and urban bridges. To ensure the accuracy of target identification results, the remote sensing images to be tested should have high temporal resolution (coverage period < 7 days), and the satellite monitoring time range should be no less than 30 days.

[0044] Please continue to refer to Figures 2-5 As shown, step S202 involves obtaining a bridge training sample set and a standard heavy-duty vehicle image training sample set for the area under study based on the obtained bridge and standard heavy-duty vehicle feature information through manual visual interpretation. This includes:

[0045] Step S2021: Based on bridge feature information, obtain typical bridge images containing lane number information within the area to be studied as a bridge training sample set by manual visual interpretation.

[0046] In this embodiment, it should be noted that typical urban bridge images containing lane number information, obtained through manual visual interpretation from the remote sensing image set of the sample study area, are used as the bridge training sample set for the area to be studied. Figure 6 As shown, the bridges in the area to be studied should be beam bridges, that is, single-span or multi-span bridges with consistent span standards, using beams or truss beams as the main load-bearing structures.

[0047] Step S2022: Based on the characteristic information of standard load vehicles, perform manual visual interpretation and combine prior knowledge and vehicle management data to obtain images of standard load vehicles of different levels in the area to be studied, which are used as training sample sets of standard load vehicle images. Among them, standard load vehicles include Class A standard load vehicles and Class B standard load vehicles.

[0048] In this embodiment, it should be noted that, in addition, remote sensing images of actual large parking lots, such as the parking areas of car dealerships, are extracted from the remote sensing image set of the sample study area, and standard load vehicle images are obtained based on manual visual interpretation as a training sample set.

[0049] According to industry standards, urban bridges are classified into two categories: "Urban-A" and "Urban-B" based on their capacity and design standards. Urban-A bridges typically have a large capacity, accommodating a large number of vehicles and pedestrians, and often involve important nodes in the urban transportation network; their smoothness directly affects the smoothness and efficiency of urban traffic. Urban-B bridges are usually used to connect different areas, streets, or important facilities within a city, serving as major secondary traffic arteries or passageways connecting important areas. Different bridge categories allow the passage of vehicles with corresponding load capacities. Therefore, by acquiring images of standard-load vehicles (Urban-A and Urban-B) and combining them with prior knowledge and vehicle management data, a training sample set was created. This training sample set includes training samples of standard-load vehicles for both Urban-A and Urban-B categories.

[0050] Specifically, such as Figure 7As shown, by combining prior knowledge and vehicle management data, design standard data for heavy-duty trucks corresponding to Class A and Class B bridge design standards are obtained, including standard truck characteristic information such as length, width, height, and actual load. Based on the obtained truck information standards, the standard load truck images obtained from remote sensing vehicle images are divided into Class A standard heavy-duty truck training sample sets and Class B standard heavy-duty truck training sample sets according to their length, width, height, and other truck characteristic information standards.

[0051] Reference Figure 2 , Figure 3 As shown, step S203, training the deep learning model based on the training sample set to construct the recognition model, includes: using the training sample set as input data based on the U-Net model to train and optimize the model in order to construct the recognition model.

[0052] In this embodiment, it should be noted that the deep learning model adopted is the U-Net model, which is a convolutional neural network model mainly used for image recognition and segmentation tasks, capable of extracting local features and spatial relationships of images and videos. The model dataset contains only positive samples and undergoes data augmentation training through steps such as random cropping, random inversion, random scaling, and random color transformation. A sliding window and hyperparameters need to be set according to the remote sensing image to be studied, and an effective algorithm should be selected for target recognition. It consists of two parts: an encoder Q and a decoder Q, connected by a skip connection. The encoder is responsible for downsampling and feature extraction. It gradually reduces the size of the feature map of the input image while increasing the abstraction of the features through a series of convolution and max pooling operations. After each downsampling, the size of the feature map is halved, and the number of channels is doubled. The decoder is used for image size restoration and segmentation. It restores the size of the feature map to the original input size through transposed convolution and convolutional layers, and uses convolutional layers to extract high-level features of the input image. The decoder's structure is symmetrical to the encoder, gradually increasing the size of the feature map through upsampling operations while reducing the number of channels. Skip connections concatenate a cropped feature map from the encoder with the corresponding feature map from the decoder. This type of connection helps fuse image features of the same size from the same layer, enabling the decoder to better recover the image's detailed information.

[0053] The U-Net model is input with three classes of training sample datasets. During training, the model preprocesses the input images (training sample sets) such as normalization and resizing. To ensure model training accuracy, the binary cross-entropy loss function is combined with the Dice loss function. This combination is not only more sensitive to boundary sharpness and the model's prediction accuracy in small regions, but also makes gradient calculation relatively easy, which is beneficial for the model's training and optimization process. In addition, the training process combines multiple optimizers and learning strategies, continuously adjusting optimization parameters and selecting the optimal solution to determine the final recognition model.

[0054] Reference Figure 2 , Figure 3 As shown, in step S204, the beam bridges in the study area and the corresponding levels of standard heavy-duty vehicles are initially identified and matched based on the recognition model, so as to determine the specific level information of the beam bridges in the study area and record their image spatial location information.

[0055] In this embodiment, it should be noted that a trained deep learning model is used to initially identify bridges and standard heavy-duty vehicles within the study area over a specific time period. This is to determine the specific grade information of beam bridges in the study area and record their spatial location information. To ensure the accuracy of the target identification results, the remote sensing images to be tested should have high temporal resolution (coverage period < 7 days) and a satellite monitoring time range of no less than 30 days. The identified beam bridges and standard heavy-duty vehicles are then matched to obtain the spatial feature information of the target objects.

[0056] A preliminary identification and matching of beam bridges and standard heavy-duty trucks within the study area was conducted, such as... Figure 8 As shown, the first step is to determine whether there is a beam bridge in the image to be studied. If so, the time attribute of the identified beam bridge is verified and the construction time range of the bridge is determined so as to match the calculation formula in the bridge design standard determined within the corresponding time period. Otherwise, the image is discarded, or a standard heavy-duty vehicle is selected again to extract ground feature information and the deep learning model is retrained.

[0057] The process of verifying and identifying the temporal attributes of beam bridges begins by confirming the presence of a beam bridge in the image under study. If no beam bridge is found, images are selected at specific time intervals until a beam bridge is identified, at which point the time of the selected image is obtained. Based on the obtained image time, the time range of the beam bridge's design or construction is determined to match the calculation formulas in the bridge design standards established within that time period. Bridge design standards are continuously updated and optimized over time, and each set of standards has different requirements for standard heavy-duty vehicles. Therefore, urban bridge design standards must be selected based on the image's temporal scale to match the bridge design standards established within the corresponding time period.

[0058] Specifically, these include standards such as the 1993 "Urban Bridge Design Guidelines" (CJJ 11-1993), the 1998 "Urban Bridge Design Load Standards" (CJJ 77-1998), the 2011 "Urban Bridge Design Specifications" (CJJ 11-2011), and the current "Highway Bridge Design Specifications" (GB 50010) and "Urban Road Bridge Design Specifications" (JTGD J11-2005 / 2014). Each set of standards has slightly different provisions for standard truck loads.

[0059] In one example, image A, dated June 2001, is selected from a remote sensing image database. No target bridge is found in image A. Next, image B, dated 2002, is selected from the database for target identification, and so on, identifying images C, D, etc. The image selection time interval can be one year, two years, or as needed, until the target bridge is identified in image N. Assuming image N was acquired in March 2010, if a target bridge is identified, its construction date is determined to be up to 2010. At this point, the relevant bridge construction specifications can be matched for further calculations.

[0060] Secondly, based on the determination that there is a beam bridge, it is further determined whether there are Class A standard heavy-duty trucks and / or Class B standard heavy-duty trucks in the image under study. If not, the image is discarded; if so, if a beam bridge and a Class A standard heavy-duty truck appear in the image at the same time, the bridge is classified as a Class A bridge; if a beam bridge, a Class A standard heavy-duty truck, and a Class B standard heavy-duty truck appear in the image at the same time, the bridge is classified as a Class A bridge; if a beam bridge and a Class B standard heavy-duty truck appear in the image at the same time, the bridge is classified as a Class B bridge.

[0061] In this embodiment, it should be noted that Class A bridges are identified first. If it is not a Class A bridge, then a Class B bridge is identified, because the load-bearing capacity of Class A bridges is superior to that of Class B bridges. If both Class A and Class B standard heavy-duty vehicles are present, it means that the bridge is at least a Class A bridge.

[0062] Finally, the spatial location information of the beam bridge, namely latitude and longitude, is recorded and verified by judging and verifying the time attributes, so as to locate the target bridge and facilitate the next step of using drones to obtain the bridge span.

[0063] like Figure 2 , Figure 3 As shown, in step S205, based on the target bridge obtained through preliminary matching, the remote sensing image of the target bridge is obtained by the UAV according to the recorded image spatial location information, so as to obtain the basic information of the target bridge.

[0064] In this embodiment, it should be noted that a multi-rotor UAV is selected to photograph the beam bridge identified through historical time verification, acquiring two-dimensional side-view images of the target bridge within the study area, and performing image processing, including image correction and image enhancement; based on the scaling ratio of the UAV remote sensing images, the net distance between two adjacent piers (or abutments) at the design flood level of the target bridge is calculated according to the distance in the image, i.e., the net span of the target bridge; simultaneously, the number of lanes and piers is obtained through manual visual interpretation, thus obtaining the basic information of the target bridge, such as... Figure 6 As shown, the basic information includes the bridge classification, bridge net span, number of piers, and number of lanes within the area to be studied.

[0065] like Figure 2 , Figure 3 As shown, step S206 involves estimating the lane load of the target bridge based on the basic information of the target bridge and according to the bridge design standards.

[0066] In this embodiment, it should be noted that the basic information of the target bridge is used to estimate the lane load of the target bridge by substituting it into the bridge lane load calculation formula in the corresponding bridge design standard. The calculation formula is expressed as: Total vehicle load of the target bridge = number of lanes × single lane load × load coefficient × number of spans.

[0067] Among them, such as Figure 9 As shown, the vehicle load on the bridge is divided into "concentrated load (P)". k ")" and "uniformly distributed load (q)" k Concentrated load refers to a locally concentrated load acting on a bridge, usually represented by a point load or a load concentrated in a small area, such as a moving vehicle, pedestrian, or equipment. Uniformly distributed load refers to a load that is evenly distributed on the bridge structure. Unlike concentrated load, uniformly distributed load is a load distributed evenly along a certain length or area. Bending moment refers to the ability or result of a material or structure to bend when subjected to external forces. When an object is subjected to lateral forces or bending moments, bending occurs, and the resulting moment is called bending moment. Bending moment is used to describe the bending state and stress conditions at various points in a structure, and is usually a moment along the axis of the cross-section. Shear force refers to a force acting on a structure or material, producing a shear effect along the direction parallel to the cross-section of the structure. In structures such as beams or bridges, shear force is a force along the cross-section parallel to the plane of force, and can be used to characterize the shear force in the stress state of the structure.

[0068] The following is a reference example of the standard and estimation formula for the single-lane load of urban bridges:

[0069] Table 1. Standard load for single lane of urban bridges and estimation formula in the "Design Load Standard for Urban Bridges" (CJJ 77-1998):

[0070] Table 1

[0071]

[0072] Table 2. Standard load and estimation formula for single lane of urban bridge in the "Code for Design of Urban Bridges" (CJJ 11-2011):

[0073] Table 2

[0074]

[0075] The implementation principle of this embodiment is as follows: The above steps mainly involve collecting remote sensing images of the area under study and combining them with prior knowledge and vehicle management data to obtain bridge and vehicle feature information within the area under study. This results in a training sample set, where vehicle loads are categorized into Class A and Class B standard trucks to match different bridge grades. Therefore, the training sample set is divided into a bridge training sample set, a Class A standard truck training sample set, and a Class B standard truck training sample set. A recognition model is constructed based on the categorized training sample set using a deep learning algorithm. This model identifies and matches beam bridges with vehicle grades, determining whether the beam bridge is a Class A or Class B bridge. This achieves the identification and extraction of urban bridges in a specific area, solving the problem of missing databases for urban bridges and their standard truck loads. Furthermore, UAV remote sensing imagery is used to supplement and verify satellite remote sensing imagery to obtain basic information such as the span, length, and width of the target bridge. Based on the bridge design standards corresponding to the target bridge, the lane load capacity of urban bridges can be effectively estimated. This solves the problem of estimating bridge vehicle loads based on remote sensing imagery when data is scarce, and provides technical support for obtaining road traffic information and route planning in special and complex geographical environments, thereby improving work efficiency.

[0076] Reference Figure 10 As shown, this application also provides a system for estimating the lane load of a target bridge based on remote sensing image detection. This system may include: an acquisition module 301, a training sample module 302, a model construction module 303, a matching module 304, a span extraction module 305, and a load estimation module 306. The main functions of each component module are as follows:

[0077] The acquisition module 301 is used to acquire a set of remote sensing images of the area to be studied in order to extract feature information of bridges and their standard load vehicles.

[0078] Training sample module 302 is used to obtain a bridge training sample set and a standard heavy-duty vehicle image training sample set for the area to be studied based on the obtained bridge and standard heavy-duty vehicle feature information and through manual visual interpretation.

[0079] The model building module 303 is used to train the deep learning model based on the training sample set and build the recognition model.

[0080] The matching module 304 is used to perform preliminary identification and matching between the beam bridges in the study area and the corresponding level of standard heavy-duty vehicles in the study area based on the recognition model, so as to determine the specific level information of the beam bridges in the study area and record their image spatial location information.

[0081] The span extraction module 305 obtains the basic information of the target bridge by using a drone to acquire remote sensing images of the target bridge based on the spatial location information of the recorded images, based on the target bridge obtained through preliminary matching. The basic information is used to display the basic construction data of the bridge.

[0082] The load estimation module 306 is used to estimate the lane load of the target bridge based on the basic information of the target bridge and according to the bridge design standards.

[0083] like Figure 11 The diagram shown is a block diagram of a computer device according to an embodiment of this application. The term "computer device" is intended to represent various forms of digital computers or mobile devices. The digital computer may include a desktop computer, a portable computer, a workbench, a personal digital assistant, a server, a mainframe computer, and other suitable computers. The mobile device may include a tablet computer, a smartphone, a wearable device, etc.

[0084] like Figure 11 As shown, device 600 includes a computing unit 601, a ROM 602, a RAM 603, a bus 604, and an input / output (I / O) interface 605. The computing unit 601, ROM 602, and RAM 603 are interconnected via the bus 604. The input / output (I / O) interface 605 is also connected to the bus 604.

[0085] The computing unit 601 can execute various processes in the method embodiments of this application according to computer instructions stored in the read-only memory (ROM) 602 or computer instructions loaded from the storage unit 608 into the random access memory (RAM) 603. The computing unit 601 can be various general-purpose and / or special-purpose processing components with processing and computing capabilities. The computing unit 601 can include, but is not limited to, a central processing unit (CPU), a graphics processing unit (GPU), various dedicated artificial intelligence (AI) computing chips, various computing units running machine learning model algorithms, a digital signal processor (DSP), and any suitable processor, controller, microcontroller, etc. In some embodiments, the methods provided in the embodiments of this application can be implemented as computer software programs, which are tangibly contained in a computer-readable storage medium, such as the storage unit 608.

[0086] RAM 603 may also store various programs and data required for the operation of device 600. Part or all of the computer program may be loaded and / or installed on device 600 via ROM 602 and / or communication unit 609.

[0087] The input unit 606, output unit 607, storage unit 608, and communication unit 609 in device 600 can be connected to I / O interface 605. The input unit 606 can be, for example, a keyboard, mouse, touchscreen, or microphone; the output unit 607 can be, for example, a display, speaker, or indicator light. Device 600 can exchange information and data with other devices through the communication unit 609.

[0088] It should be noted that the device may also include other components necessary for normal operation. It may also include only the components necessary for implementing the solution of this application, without necessarily including all the components shown in the figures.

[0089] Various implementations of the systems and techniques described herein can be implemented in digital electronic circuit systems, integrated circuit systems, field programmable gate arrays (FPGAs), application-specific integrated circuits (ASICs), application-specific standard products (ASSPs), systems-on-a-chip (SOCs), payload programmable logic devices (CPLDs), computer hardware, firmware, software, and / or combinations thereof.

[0090] The computer instructions used to implement the methods of this application may be written in any combination of one or more programming languages. These computer instructions may be provided to the computing unit 601 such that when executed by the computing unit 601, such as a processor, the computer instructions cause the execution of the steps involved in the embodiments of the methods of this application.

[0091] The computer-readable storage medium provided in this application can be a tangible medium that can contain or store computer instructions for performing the steps involved in the method embodiments of this application. The computer-readable storage medium can be, but is not limited to, electronic, magnetic, optical, electromagnetic, and other forms of storage media.

[0092] The specific embodiments described above do not constitute a limitation on the scope of protection of this application. Those skilled in the art should understand that various modifications, combinations, sub-combinations, and substitutions can be made according to design requirements and other factors. Any modifications, equivalent substitutions, and improvements made within the spirit and principles of this application should be included within the scope of protection of this application.

Claims

1. A method for estimating the lane load of a target bridge based on remote sensing image detection, characterized in that, Includes the following steps: Collect remote sensing image sets of the area to be studied to extract characteristic information of bridges and their standard load vehicles; Based on the obtained bridge and standard load vehicle feature information, a bridge training sample set and a standard load vehicle image training sample set for the area to be studied are obtained by manual visual interpretation. The deep learning model is trained based on the training sample set to construct a recognition model; Based on the aforementioned recognition model, preliminary identification and matching are performed between beam bridges and standard heavy-duty trucks in the study area to determine the specific grade information of beam bridges in the study area and record their image spatial location information, including: Determine whether there is a beam bridge in the image to be studied. If so, verify the time attribute of the identified beam bridge and determine the bridge construction time range to match the calculation formula in the bridge design standard determined within the corresponding time period. Otherwise, discard the image. Based on the determination of a beam bridge, further determine whether there are Class A standard heavy-duty trucks and / or Class B standard heavy-duty trucks in the image to be studied. If not, discard the image. If both a beam bridge and a Class A standard heavy-duty truck appear simultaneously in the image, the bridge is classified as a Class A bridge. If a beam bridge, a Class A standard heavy-duty truck, and a Class B standard heavy-duty truck all appear in the image at the same time, then the bridge is classified as a Class A bridge. If both a beam bridge and a Class B standard heavy-duty truck appear in the image, the bridge is classified as a Class B bridge. Record the spatial location information of the beam bridge by identifying and verifying the time attribute; Based on the target bridge obtained through preliminary matching, remote sensing images of the target bridge are obtained by UAV according to the recorded spatial location information of the images, so as to obtain the basic information of the target bridge. The basic information is used to display the basic construction data of the target bridge, including the bridge classification, bridge clear span, number of piers, and number of lanes in the area to be studied. Based on the basic information of the target bridge, the lane load of the target bridge is estimated according to the bridge design standards, including: The obtained basic information of the target bridge is used to estimate the lane load of the target bridge by substituting it into the bridge lane load calculation formula in the corresponding bridge design standard. The calculation formula is expressed as follows: Total vehicle load of the target bridge = number of lanes × single lane load × load factor × number of spans.

2. The method for estimating the lane load of a target bridge based on remote sensing image detection according to claim 1, characterized in that, The collection of remote sensing images of the area under study is used to extract feature information of bridges and their standard load vehicles, including: Collect real bridge construction environment data within the study area and obtain corresponding remote sensing images; The remote sensing images are preprocessed, and a set of remote sensing images of the area to be studied is obtained by cropping based on visual interpretation, so as to extract the characteristic information of bridges and standard load vehicles in the area to be studied.

3. The method for estimating the lane load of a target bridge based on remote sensing image detection according to claim 2, characterized in that, Based on the obtained bridge and standard loaded vehicle feature information, and through manual visual interpretation, a bridge training sample set and a standard loaded vehicle image training sample set for the area under study are obtained, including: Based on bridge feature information, typical bridge images containing lane number information are obtained from the area to be studied through manual visual interpretation and used as a bridge training sample set. Based on the characteristic information of standard load vehicles, manual visual interpretation is performed, and images of standard load vehicles of different levels in the area to be studied are obtained by combining prior knowledge and vehicle management data. These images are used as training sample sets for standard load vehicles. The standard load vehicles include Class A standard load vehicles and Class B standard load vehicles.

4. The method for estimating the lane load of a target bridge based on remote sensing image detection according to claim 3, characterized in that, The step of training the deep learning model based on the training sample set to construct the recognition model includes: The U-Net model is used as input data to train and optimize the model in order to build a recognition model.

5. The method for estimating the lane load of a target bridge based on remote sensing image detection according to claim 1, characterized in that, The process involves determining whether a beam bridge exists within the image under study. If so, the temporal attribute of the identified beam bridge is verified to determine the bridge's construction time range, in order to match the calculation formula in the bridge design standards determined within the corresponding time period. This specifically includes: Based on the images to be studied, first confirm whether there is a beam bridge. If not, continue to select images to be studied according to the time interval as needed to confirm whether there is a beam bridge. If so, obtain the image time of the image to be studied; Based on the acquired image time, the time range of the beam bridge during its design or construction is determined to match the calculation formula in the bridge design standard determined within the corresponding time period. The bridge design standard is continuously updated and optimized over time, and each set of bridge design standards has different provisions for standard heavy-duty trucks.

6. The method for estimating the lane load of a target bridge based on remote sensing image detection according to claim 1, characterized in that, The target bridge, obtained based on preliminary matching, is used by an unmanned aerial vehicle (UAV) to acquire remote sensing images of the target bridge based on the recorded spatial location information of the images, in order to obtain basic information about the target bridge, including: Two-dimensional side-view images of the target bridge within the area to be studied are acquired and image processing is performed. The net span of the target bridge is calculated based on the distance in the image according to the scaling ratio of the UAV remote sensing image. At the same time, the number of lanes and piers is obtained through manual visual interpretation to obtain the basic information of the target bridge. The basic information includes the bridge classification, net span, number of piers, and number of lanes in the area to be studied.

7. A system for estimating the lane load of a target bridge based on remote sensing imagery, characterized in that, include: The acquisition module is used to collect remote sensing image sets of the area to be studied in order to extract feature information of bridges and their standard load vehicles. The training sample module is used to obtain a bridge training sample set and a standard heavy-duty vehicle image training sample set for the area under study based on the obtained feature information of the bridge and the standard heavy-duty vehicle, according to manual visual interpretation. A model building module is used to train a deep learning model based on the training sample set and build a recognition model; The matching module is used to perform preliminary identification and matching between beam bridges in the study area and the corresponding levels of standard heavy-duty vehicles in the study area based on the recognition model, so as to determine the specific level information of beam bridges in the study area and record their image spatial location information, including: Determine whether there is a beam bridge in the image to be studied. If so, verify the time attribute of the identified beam bridge and determine the bridge construction time range to match the calculation formula in the bridge design standard determined within the corresponding time period. Otherwise, discard the image. Based on the determination of a beam bridge, further determine whether there are Class A standard heavy-duty trucks and / or Class B standard heavy-duty trucks in the image to be studied. If not, discard the image. If both a beam bridge and a Class A standard heavy-duty truck appear simultaneously in the image, the bridge is classified as a Class A bridge. If a beam bridge, a Class A standard heavy-duty truck, and a Class B standard heavy-duty truck all appear in the image at the same time, then the bridge is classified as a Class A bridge. If both a beam bridge and a Class B standard heavy-duty truck appear in the image, the bridge is classified as a Class B bridge. Record the spatial location information of the beam bridge by identifying and verifying the time attribute; The span extraction module is used to obtain remote sensing images of the target bridge based on the preliminary matching target bridge by using a UAV according to the recorded image spatial location information, so as to obtain the basic information of the target bridge. The basic information is used to display the basic construction data of the target bridge, including bridge classification, bridge net span, number of piers, and number of lanes in the area to be studied. The load estimation module is used to estimate the lane load of the target bridge based on the basic information of the target bridge and according to bridge design standards, including: The obtained basic information of the target bridge is used to estimate the lane load of the target bridge by substituting it into the bridge lane load calculation formula in the corresponding bridge design standard. The calculation formula is expressed as: Total vehicle load of the target bridge = Number of lanes × Single lane load × Load factor × Number of spans.

8. A computer device, comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that, When the processor executes the computer program, it implements the steps of the method according to any one of claims 1 to 6.

Images