A Bridge Identity Intelligent Recognition Method Based on Physical Features

Abstract

This invention discloses a bridge identification intelligent method based on physical features. It utilizes photogrammetric three-dimensional reconstruction and spatial shape measurement technology for each bridge component to identify the components of bridges under construction and those in operation. The identification is based on the time-varying physical feature group of bridge components.<G,*> The intelligent bridge identification method mainly includes a storage module and a processing module. The storage module stores all information about each component of the bridge, including the physical feature groups from each 3D photographic reconstruction.<G,*> It can quickly output the key parameters of the components of interest; the processing module mainly performs all information data tracing and calculation processing of 3D photogrammetry reconstruction. Based on the parameter recognition method of normal vector region growth, it realizes ID recognition and encoding storage of faces, edges, and lines in the surface model corresponding to the point cloud model of 3D bridge components. It can realize anti-counterfeiting recognition and traceability of various components of bridge engineering, and has the characteristics of high precision, non-contact, intelligence, and speed and efficiency.

Description

Technical Field

[0001] This application relates to the field of intelligent control technology, and in particular to a method for intelligent identification of bridge identity based on physical characteristics. Background Technology

[0002] Currently, in my country's bridge maintenance and management, information on the safety and health status of bridges is mainly obtained through visual inspection and manual measurement methods, using measuring devices such as rulers and tape measures. This not only fails to comprehensively reflect the dimensional characteristics of components but also suffers from drawbacks such as high workload, low accuracy, low efficiency, and limited measurement data. Manual inspection is also difficult to use for real-time health checks of large bridges. Furthermore, during bridge operation, RFID chips can be embedded in the beam components or QR code nameplates can be placed on the surface to store and identify component information. However, RFID radio frequency signals gradually weaken, and the lifespan of RFID chips and sensors is approximately 5-10 years. Given that the bridge's lifespan is 100 years, the RFID chips and sensors within the beam components need to be replaced periodically, increasing the later-stage bridge operation and maintenance costs. QR code nameplates, which are exposed to the surface of components for extended periods, are easily damaged by external factors and also require regular maintenance or replacement.

[0003] To improve the management efficiency of bridge engineering and strengthen the quality control and supervision of bridges, a bridge identity intelligent recognition method based on physical characteristics is proposed. Summary of the Invention

[0004] In order to solve the above-mentioned technical problems, the present invention provides a bridge identity intelligent recognition method based on physical features.

[0005] The objective of this invention can be achieved through the following technical solutions:

[0006] A bridge identification intelligent method based on physical features is proposed. This method utilizes photogrammetric 3D reconstruction and spatial shape measurement technology for each bridge component to identify the components of bridges under construction and in operation. The identification is based on the physical feature groups of the bridge components.<G,*> Among them, the feature group<G,*> The characterization elements include: coordinate data and feature data such as special marker points, external dimensions, apparent color, and cracks, specifically characterized as follows:<G,*> = {label data, shape data, coordinate data, RGB color values, normal vector, characteristic defect data, ...}; the physical feature group<G,*> The following relationship must be satisfied:

[0007]

[0008]

[0009] ...

[0010]

[0011] ...

[0012] In the formula:<G,*> t＝i —The physical feature group of bridge components after photogrammetry and 3D reconstruction at t=i;

[0013] —Time-varying correction coefficient, satisfying Where Δt i =t i -t i-1 .

[0014] The physical feature group<G,*> t＝i Physical feature group of bridge components after the previous two photogrammetry and 3D reconstruction<G,*> t＝i-1 and<G,*> t＝i-2 They are related and satisfy the following relationship:

[0015]

[0016]

[0017] ...

[0018]

[0019] ...

[0020] Where: μ i —Time-varying proportional coefficient, satisfying μ i =f(Δt) i ), where Δt i =t i -t i-1 .

[0021] The bridge identification intelligent recognition method mainly includes a storage module and a processing module; the storage module stores all information of each bridge component, including the physical feature group of each 3D photographic reconstruction.<G,*> It can quickly output key parameters of the components of interest (such as length, width, height, angle, point distance, and changes in defect characteristics); the processing module mainly performs all information data tracing and calculation processing of 3D photogrammetry reconstruction. Based on the parameter recognition method of normal vector region growth, it can automatically identify and process the spatial distribution characteristics of point cloud data. Through the differences in the normal vectors of micro-region points on the surface of different 2D images and the topological relationship between points, lines, and surfaces of the component shape, it realizes ID recognition and encoding storage of surfaces, edges, and lines in the surface model corresponding to the point cloud model of the 3D bridge component.

[0022] The physical characteristics of the bridge components<G,*> It is time-varying, and its acquisition is achieved through photogrammetry, which captures and measures features of the bridge structure such as marker points, dimensions, deformation, deflection, shape changes, cracks, and exposed reinforcement. The captured two-dimensional images are then imported into a computer information management platform, where the spatial coordinates of the captured image point cloud are calibrated and matched with the actual coordinates. The captured bridge is then reconstructed in three dimensions. The physical feature data of the entire three-dimensional photogrammetric reconstruction technology constitutes a physical feature group.<G,*> Furthermore, the physical feature clusters formed by the first 3D photographic reconstruction are<G,*> t＝0 , and thereafter, are represented as follows:<G,*> t＝1 ,<G,*> t＝2 ...<G,*> t＝i ...; the dimensional detection accuracy of the aforementioned three-dimensional photogrammetric reconstruction technology can reach 2mm.

[0023] The aforementioned marker points are obvious characteristic defects (such as cracks, exposed rebar, large-area concrete spalling, etc.) at special locations on bridge components. During image recognition, the computer first identifies and sets coded points for obvious and special marker points, then extracts their features, separates them from the two-dimensional image point cloud through convolution operations, calculates the center coordinates of the marker points using a quadratic curve fitting algorithm, and finally decodes the numbers represented by the marker points according to the encoding rules. Then, multiple two-dimensional images are registered based on overlapping parts and transformed to the same coordinate system, thereby registering them one by one with other actual marker points. The point cloud information of the two-dimensional image has nine data representations, including point spatial coordinates XYZ, RGB color values, and normal vectors.

[0024] The beneficial effects achieved by this application are as follows:

[0025] 1. Bridge identification based on 3D photogrammetry reconstruction technology features high precision, non-contact operation, intelligence, automation, and high speed and efficiency;

[0026] 2. A high-precision three-dimensional reconstruction model of the entire bridge can collect information and identify components of the bridge in all aspects, and establish a comprehensive evaluation system to achieve anti-counterfeiting, tracking and traceability of various components of the bridge project.

[0027] 3. Three-dimensional photography reconstruction technology can help us understand the health status of bridge components in a timely manner, detect potential safety hazards as early as possible, and carry out timely maintenance and repair of bridge projects, thereby reducing the probability of catastrophic accidents. Attached Figure Description

[0028] Figure 1 This is a schematic diagram of the bridge identity intelligent recognition method based on three-dimensional photogrammetry reconstruction technology of the present invention;

[0029] Figure 2 This is a schematic diagram of the marking points of the present invention. Detailed Implementation

[0030] The following is in conjunction with the appendix Figures 1-2 The present invention will be described in detail below with specific embodiments. These embodiments are based on the technical solutions of the present invention and provide detailed implementation methods and specific operating procedures. However, the scope of protection of the present invention is not limited to the following embodiments.

[0031] A bridge identification intelligent method based on physical features utilizes photogrammetric 3D reconstruction and spatial shape measurement technology for each bridge component to identify the components of bridges under construction and in operation; for example... Figure 1 As shown, the identity recognition is based on the physical feature group of bridge components.<G,*> Among them, the feature group<G,*> The characterization elements include: coordinate data and feature data such as special marker points, external dimensions, apparent color, and cracks, specifically characterized as follows:<G,*> = {label data, shape data, coordinate data, RGB color values, normal vector, characteristic defect data, ...}; the physical feature group<G,*> The following relationship must be satisfied:

[0032]

[0033]

[0034] ...

[0035]

[0036] ...

[0037] In the formula:<G,*> t＝i —The physical feature group of bridge components after photogrammetry and 3D reconstruction at t=i;

[0038] —Time-varying correction coefficient, satisfying Where Δt i =t i -t i-1 .

[0039] The physical feature group<G,*> t＝i Physical feature group of bridge components after the previous two photogrammetry and 3D reconstruction<G,*> t＝i-1 and<G,*> t＝i-2 They are related and satisfy the following relationship:

[0040]

[0041]

[0042] ...

[0043]

[0044] ...

[0045] Where: μ i —Time-varying proportional coefficient, satisfying μ i =f(Δt) i ), where Δt i =t i -t i-1 .

[0046] The bridge identification intelligent recognition method mainly includes a storage module and a processing module; the storage module stores all information of each bridge component, including the physical feature group of each 3D photographic reconstruction.<G,*> It can quickly output key parameters of the components of interest (such as length, width, height, angle, point distance, and changes in defect characteristics); the processing module mainly performs all information data tracing and calculation processing of 3D photogrammetry reconstruction. Based on the parameter recognition method of normal vector region growth, it can automatically identify and process the spatial distribution characteristics of point cloud data. Through the differences in the normal vectors of micro-region points on the surface of different 2D images and the topological relationship between points, lines, and surfaces of the component shape, it realizes ID recognition and encoding storage of surfaces, edges, and lines in the surface model corresponding to the point cloud model of the 3D bridge component.

[0047] The physical characteristics of the bridge components<G,*> It is time-varying, and its acquisition is achieved through photogrammetry, which captures and measures features of the bridge structure such as marker points, dimensions, deformation, deflection, shape changes, cracks, and exposed reinforcement. The captured two-dimensional images are then imported into a computer information management platform, where the spatial coordinates of the captured image point cloud are calibrated and matched with the actual coordinates. The captured bridge is then reconstructed in three dimensions. The physical feature data of the entire three-dimensional photogrammetric reconstruction technology constitutes a physical feature group.<G,*> Furthermore, the physical feature clusters formed by the first 3D photographic reconstruction are<G,*> t＝0 , and thereafter, are represented as follows:<G,*> t＝1 ,<G,*> t＝2 ...<G,*> t＝i ...; the dimensional detection accuracy of the aforementioned three-dimensional photogrammetric reconstruction technology can reach 2mm.

[0048] like Figure 2As shown, the marker points are obvious characteristic defects at special locations on bridge components. For example, taking a bridge T-beam as an example, obvious cracks and exposed rebar at the mid-span of the T-beam can be designated as marker point 1 and marker point 2, respectively. During image recognition, the computer first identifies and sets an encoding point for the obviously special marker point 1, then extracts its crack features (crack length, width, direction, etc.), separates it from the two-dimensional image point cloud through convolution operation, calculates the center coordinates of the marker point using a quadratic curve fitting algorithm, and finally decodes the numbers represented by the marker points according to the encoding rules. Then, multiple two-dimensional images are registered based on overlapping parts and transformed to the same coordinate system, thereby registering them one by one with other actual marker points. The point cloud information of the two-dimensional image has nine data representations, including point spatial coordinates XYZ, RGB color values, and normal vectors.

[0049] The above description of the embodiments is provided to enable those skilled in the art to understand and use them. It will be apparent to those skilled in the art that various modifications can be easily made to these embodiments, and the general principles described herein can be applied to other embodiments without inventive effort. Therefore, the present invention is not limited to the above embodiments, and any improvements and modifications made by those skilled in the art based on the disclosure of this patent, without departing from the scope of the present invention, should be within the protection scope of this patent.

Claims

1. A bridge identity intelligent recognition method based on physical features, characterized in that, Based on photogrammetric 3D reconstruction and spatial shape measurement technology for bridge components, this method identifies the components of bridges under construction and in operation; the identification is based on the physical feature groups of the bridge components. Among them, the feature group The characterization elements include: marker points, external dimensions, and apparent color, specifically: {Shape data, coordinate data, RGB color values, normal vector, characteristic defect data}; the aforementioned physical feature group The following relationship must be satisfied: ； ； …… ； …… In the formula: —— At that time, the physical feature group of bridge components after photogrammetry and 3D reconstruction; —Time-varying correction coefficient, satisfying ,in ; The physical feature group Physical feature group of bridge components after the previous two photogrammetry and 3D reconstruction and They are related and satisfy the following relationship: ； ； …… ； …… In the formula: —Time-varying proportional coefficient, satisfying ,in ; The bridge identification intelligent recognition method includes a storage module and a processing module; the storage module stores all information of each bridge component, including the physical feature group of each 3D photographic reconstruction. The processing module performs all information data tracing and calculation processing for 3D photogrammetry reconstruction. Based on the parameter identification method of normal vector region growth, it identifies and processes the spatial distribution characteristics of point cloud data. Through the differences in the normal vectors of micro-region points on the surface of different 2D images and the topological relationship between points, lines and surfaces of the component shape, it performs ID identification and encoding storage of the surfaces, edges and lines in the surface model corresponding to the 3D bridge component point cloud model. The physical characteristics of the bridge components The data is obtained by taking photos of the bridge structure's marker points, dimensions, deformation, deflection, and shape changes using photogrammetry. The captured two-dimensional images are then imported into a computer information management platform for point cloud spatial coordinate calibration. After calibration, these coordinates are calculated and matched with the actual coordinates to perform three-dimensional reconstruction of the bridge. The physical feature data from the entire three-dimensional photogrammetric reconstruction technology constitute a physical feature group. ; The marker points are characteristic defects of bridge components. During image recognition, the computer first identifies and sets encoding points for the marker points, then extracts their features, separates them from the two-dimensional image point cloud through convolution operations, calculates the center coordinates of the marker points using a quadratic curve fitting algorithm, and finally decodes the numbers represented by the marker points according to the encoding rules. Then, multiple two-dimensional images are registered based on overlapping parts and transformed to the same coordinate system, thereby registering them one by one with other actual marker points.

Images