Bridge construction monitoring method and system based on AI platform

Abstract

The invention provides a bridge construction monitoring method and system based on an AI platform, and the method comprises the steps: carrying out the photographing of a bridge construction site from a first angle and a second angle, obtaining a corresponding bridge construction image, analyzing the corresponding bridge construction image, obtaining the existence state information of a first bridge member and the existence state information of a second bridge member, and carrying out the analysis of the information. Carrying out multi-view positioning identification on the bridge member; aI analysis is carried out on the existence state information of the first bridge component and the existence state information of the second bridge component to obtain first pose information and second pose information corresponding to the same component, so that whether the same bridge component is in a normal pose state or not is judged, and the accuracy of detecting the pose state of the bridge component is improved; and when the same bridge component is not in the normal pose state, the operation parameters for pose adjustment of the same bridge component are determined, so that the operation instruction is generated and sent to the construction equipment, and the construction equipment can conveniently and accurately adjust the pose of the bridge component in time.

Description

Technical Field

[0001] The present invention relates to the field of artificial intelligence, and in particular to a bridge construction monitoring method and system based on an AI platform. Background Art

[0002] During the construction of a bridge, components such as steel beams and cables need to be installed in the corresponding positions to ensure the stability of the main structure of the bridge. At present, total stations are usually used to locate bridge components, but total stations need to be calibrated before use, and total stations can only locate and detect a single bridge component at the same time, and cannot quickly and accurately locate and identify a large number of bridge components, which increases the workload and reliability of locating bridge components during bridge construction. Summary of the invention

[0003] The purpose of the present invention is to provide a bridge construction monitoring method and system based on an AI platform, which photographs the bridge construction site from a first angle and a second angle respectively to obtain corresponding bridge construction images, and analyzes the corresponding bridge construction images to obtain first bridge component existence status information and second bridge component existence status information, and performs multi-perspective positioning and identification of bridge components; performs AI analysis on the first bridge component existence status information and the second bridge component existence status information to obtain first posture information and second posture information corresponding to the same component, thereby judging whether the same bridge component is in a normal posture state, thereby improving the accuracy of bridge component posture state detection; when the same bridge component is not in a normal posture state, determining the operating parameters for adjusting the posture of the same bridge component, thereby generating operating instructions and sending them to construction equipment, so as to facilitate the construction equipment to make timely and accurate posture adjustments to the bridge components, thereby reducing the workload of bridge component positioning.

[0004] The present invention is achieved through the following technical solutions:

[0005] Bridge construction monitoring method based on AI platform, including:

[0006] Photographing the bridge construction site from a first angle to obtain a first bridge construction image; analyzing the first bridge construction image to obtain the existence status information of the first bridge component at the bridge construction site;

[0007] photographing the bridge construction site from a second angle to obtain a second bridge construction image; analyzing the second bridge construction image to obtain the second bridge component existence status information of the bridge construction site;

[0008] Uploading the first bridge component existence status information and the second bridge component existence status information to the AI ​​platform, and extracting the first posture information and the second posture information corresponding to the same bridge component; judging whether the same bridge component is in a normal posture state based on the first posture information and the second posture information;

[0009] When the same bridge component is not in a normal posture state, the operating parameters for adjusting the posture of the same bridge component are determined based on the first bridge construction image and the second bridge construction image, and corresponding operating instructions are generated based on the operating parameters and sent to the corresponding construction equipment.

[0010] Optionally, photographing the bridge construction site from a first angle to obtain a first bridge construction image; analyzing the first bridge construction image to obtain the first bridge component existence status information of the bridge construction site includes:

[0011] Perform binocular photography of the bridge construction site from a ground perspective to obtain a binocular image of the bridge construction site; and generate a first three-dimensional bridge construction image based on the binocular parallax of the binocular image;

[0012] The first bridge construction three-dimensional image is subjected to picture pixel contour recognition processing to obtain first edge contour distribution state information of all bridge components at the bridge construction site.

[0013] Optionally, photographing the bridge construction site from a second angle to obtain a second bridge construction image; analyzing the second bridge construction image to obtain the second bridge component existence status information at the bridge construction site includes:

[0014] Scan and photograph the bridge construction site from a high altitude to obtain a panoramic image of the bridge construction site; perform three-dimensional conversion on the panoramic image to obtain a second three-dimensional image of the bridge construction site;

[0015] The second bridge construction three-dimensional image is subjected to picture pixel contour recognition processing to obtain second edge contour distribution state information of all bridge components at the bridge construction site.

[0016] Optionally, scanning and photographing the bridge construction site from a high altitude angle includes:

[0017] Extract the panoramic image corresponding to each scanning shot;

[0018] Comparing the similarity of the panoramic images corresponding to each two adjacent scanning moments to obtain the similarity value between the panoramic images corresponding to each two adjacent scanning moments;

[0019] Extract the scanning time interval between every two adjacent scanning moments;

[0020] Obtaining a scanning time interval adjustment coefficient by using the scanning time interval between each two adjacent scanning moments and a similarity value between panoramic images corresponding to each two adjacent scanning moments;

[0021] The scanning time interval adjustment coefficient is obtained by the following formula:

[0022]

[0023] Wherein, K represents the scanning time interval adjustment coefficient; n represents the number of panoramic image groups corresponding to every two adjacent scanning moments; S i S represents the similarity value of the panoramic images corresponding to two adjacent scanning moments of the i-th group; i+1 T represents the similarity value of the panoramic images corresponding to two adjacent scanning moments of the i+1th group; gi T represents the scanning time interval of the panoramic image corresponding to two adjacent scanning moments of the i-th group; gi+1 represents the scanning time interval of the panoramic image corresponding to two adjacent scanning moments of the i+1th group; S min represents the minimum similarity value of the panoramic images corresponding to n groups of two adjacent scanning moments; T gmin S represents the scanning time interval corresponding to the minimum similarity value of the panoramic images corresponding to n groups of two adjacent scanning moments; b T represents the standard deviation of the similarity values ​​of the panoramic images corresponding to n groups of two adjacent scanning moments; gb It represents the standard deviation of the scanning time interval corresponding to two adjacent scanning moments of n groups;

[0024] Comparing the scanning time interval adjustment coefficient with a preset adjustment coefficient threshold;

[0025] When the scanning time interval adjustment coefficient is lower than a preset adjustment coefficient threshold, the scanning time interval is not adjusted;

[0026] When the scanning time interval adjustment coefficient is not lower than a preset adjustment coefficient threshold, retrieving the scanning time interval adjustment coefficient;

[0027] Using the scanning time interval adjustment coefficient to adjust the scanning time interval to obtain an adjusted scanning time interval;

[0028] The adjusted scanning time interval is obtained by the following formula:

[0029]

[0030] Among them, T gtRepresents the adjusted scanning time interval; T g represents the scanning time interval before adjustment; K represents the scanning time interval adjustment coefficient; K0 represents the preset adjustment coefficient threshold; S k S represents the similarity value between the corresponding panoramic image when the scanning time interval adjustment coefficient is not less than the preset adjustment coefficient threshold and the corresponding panoramic image taken in the previous scanning; p Represents the average value of similarity between n groups of panoramic images corresponding to two adjacent scanning moments;

[0031] The bridge construction site is scanned and photographed according to the adjusted scanning time interval.

[0032] Optionally, uploading the first bridge component existence status information and the second bridge component existence status information to an AI platform, extracting first posture information and second posture information corresponding to the same bridge component therefrom; and judging whether the same bridge component is in a normal posture state based on the first posture information and the second posture information, including:

[0033] Uploading the first edge contour distribution state information and the second edge contour distribution state information to the AI ​​platform, performing AI recognition on the first edge contour distribution state information and the second edge contour distribution state information, respectively, to obtain the edge contour spatial distribution information of all bridge components in the first bridge construction 3D image and the edge contour spatial distribution information of all bridge components in the second bridge construction 3D image;

[0034] Compare the edge contour spatial distribution information of all bridge components in the first bridge construction image and the edge contour spatial distribution information of all bridge components in the second bridge construction image, and extract the first placement posture information and the second placement posture information corresponding to the same bridge component;

[0035] The first placement posture information and the second placement posture information are integrated into the same coordinate space for comparison to obtain corresponding posture deviation values; if the posture deviation value is less than a preset deviation threshold, it is judged that the same bridge component is in a normal posture state; otherwise, it is judged that the same bridge component is not in a normal posture state.

[0036] Optionally, when the same bridge component is not in a normal posture state, an operation parameter for adjusting the posture of the same bridge component is determined based on the first bridge construction image and the second bridge construction image, and a corresponding operation instruction is generated based on the operation parameter, and sent to a corresponding construction equipment, including:

[0037] When the same bridge component is not in a normal posture state, the first bridge construction three-dimensional image and the second bridge construction three-dimensional image are analyzed to determine the six-degree-of-freedom adjustment parameters for adjusting the posture of the same bridge component, and corresponding operation instructions are generated based on the six-degree-of-freedom adjustment parameters. The operation instructions are encrypted and sent to the corresponding construction equipment.

[0038] The bridge construction monitoring system based on AI platform includes:

[0039] A first image shooting and analysis module is used to shoot the bridge construction site from a first angle to obtain a first bridge construction image; analyze the first bridge construction image to obtain the existence status information of the first bridge component of the bridge construction site;

[0040] A second image shooting and analysis module is used to shoot the bridge construction site from a second angle to obtain a second bridge construction image; analyze the second bridge construction image to obtain the second bridge component existence status information of the bridge construction site;

[0041] An AI processing module is used to upload the first bridge component existence state information and the second bridge component existence state information to an AI platform, and extract the first posture information and the second posture information corresponding to the same bridge component; based on the first posture information and the second posture information, determine whether the same bridge component is in a normal posture state;

[0042] The bridge component adjustment module is used to determine the operating parameters for adjusting the posture of the same bridge component based on the first bridge construction image and the second bridge construction image when the same bridge component is not in a normal posture state, and generate corresponding operating instructions based on the operating parameters, and send them to the corresponding construction equipment.

[0043] Optionally, the first image shooting and analysis module is used to shoot the bridge construction site from a first angle to obtain a first bridge construction image; and analyze the first bridge construction image to obtain the existence status information of the first bridge component at the bridge construction site, including:

[0044] Perform binocular photography of the bridge construction site from a ground perspective to obtain a binocular image of the bridge construction site; and generate a first three-dimensional bridge construction image based on the binocular parallax of the binocular image;

[0045] The first bridge construction three-dimensional image is subjected to picture pixel contour recognition processing to obtain first edge contour distribution state information of all bridge components at the bridge construction site.

[0046] Optionally, the second image shooting and analysis module is used to shoot the bridge construction site from a second angle to obtain a second bridge construction image; and analyze the second bridge construction image to obtain the second bridge component existence status information of the bridge construction site, including:

[0047] Scan and photograph the bridge construction site from a high altitude to obtain a panoramic image of the bridge construction site; perform three-dimensional conversion on the panoramic image to obtain a second three-dimensional image of the bridge construction site;

[0048] The second bridge construction three-dimensional image is subjected to picture pixel contour recognition processing to obtain second edge contour distribution state information of all bridge components at the bridge construction site.

[0049] Optionally, scanning and photographing the bridge construction site from a high altitude angle includes:

[0050] Extract the panoramic image corresponding to each scanning shot;

[0051] Comparing the similarity of the panoramic images corresponding to each two adjacent scanning moments to obtain the similarity value between the panoramic images corresponding to each two adjacent scanning moments;

[0052] Extract the scanning time interval between every two adjacent scanning moments;

[0053] Obtaining a scanning time interval adjustment coefficient by using the scanning time interval between each two adjacent scanning moments and a similarity value between panoramic images corresponding to each two adjacent scanning moments;

[0054] The scanning time interval adjustment coefficient is obtained by the following formula:

[0055]

[0056] Wherein, K represents the scanning time interval adjustment coefficient; n represents the number of panoramic image groups corresponding to every two adjacent scanning moments; S i S represents the similarity value of the panoramic images corresponding to two adjacent scanning moments of the i-th group; i+1 T represents the similarity value of the panoramic images corresponding to two adjacent scanning moments of the i+1th group; gi T represents the scanning time interval of the panoramic image corresponding to two adjacent scanning moments of the i-th group; gi+1 represents the scanning time interval of the panoramic image corresponding to two adjacent scanning moments of the i+1th group; S min represents the minimum similarity value of the panoramic images corresponding to n groups of two adjacent scanning moments; T gmin S represents the scanning time interval corresponding to the minimum similarity value of the panoramic images corresponding to n groups of two adjacent scanning moments;b T represents the standard deviation of the similarity values ​​of the panoramic images corresponding to n groups of two adjacent scanning moments; gb It represents the standard deviation of the scanning time interval corresponding to two adjacent scanning moments of n groups;

[0057] Comparing the scanning time interval adjustment coefficient with a preset adjustment coefficient threshold;

[0058] When the scanning time interval adjustment coefficient is lower than a preset adjustment coefficient threshold, the scanning time interval is not adjusted;

[0059] When the scanning time interval adjustment coefficient is not lower than a preset adjustment coefficient threshold, retrieving the scanning time interval adjustment coefficient;

[0060] Using the scanning time interval adjustment coefficient to adjust the scanning time interval to obtain an adjusted scanning time interval;

[0061] The adjusted scanning time interval is obtained by the following formula:

[0062]

[0063] Among them, T gt Represents the adjusted scanning time interval; T g represents the scanning time interval before adjustment; K represents the scanning time interval adjustment coefficient; K0 represents the preset adjustment coefficient threshold; S k S represents the similarity value between the corresponding panoramic image when the scanning time interval adjustment coefficient is not less than the preset adjustment coefficient threshold and the corresponding panoramic image taken in the previous scanning; p Represents the average value of similarity between n groups of panoramic images corresponding to two adjacent scanning moments;

[0064] The bridge construction site is scanned and photographed according to the adjusted scanning time interval.

[0065] Optionally, the AI ​​processing module is used to upload the first bridge component existence status information and the second bridge component existence status information to the AI ​​platform, and extract the first posture information and the second posture information corresponding to the same bridge component; based on the first posture information and the second posture information, judging whether the same bridge component is in a normal posture state, including:

[0066] Uploading the first edge contour distribution state information and the second edge contour distribution state information to the AI ​​platform, performing AI recognition on the first edge contour distribution state information and the second edge contour distribution state information, respectively, to obtain the edge contour spatial distribution information of all bridge components in the first bridge construction 3D image and the edge contour spatial distribution information of all bridge components in the second bridge construction 3D image;

[0067] Compare the edge contour spatial distribution information of all bridge components in the first bridge construction image and the edge contour spatial distribution information of all bridge components in the second bridge construction image, and extract the first placement posture information and the second placement posture information corresponding to the same bridge component;

[0068] The first placement posture information and the second placement posture information are integrated into the same coordinate space for comparison to obtain corresponding posture deviation values; if the posture deviation value is less than a preset deviation threshold, it is judged that the same bridge component is in a normal posture state; otherwise, it is judged that the same bridge component is not in a normal posture state.

[0069] Optionally, the bridge component adjustment module is used to determine, when the same bridge component is not in a normal posture state, operation parameters for adjusting the posture of the same bridge component based on the first bridge construction image and the second bridge construction image, and generate corresponding operation instructions based on the operation parameters, and send them to corresponding construction equipment, including:

[0070] When the same bridge component is not in a normal posture state, the first bridge construction three-dimensional image and the second bridge construction three-dimensional image are analyzed to determine the six-degree-of-freedom adjustment parameters for adjusting the posture of the same bridge component, and corresponding operation instructions are generated based on the six-degree-of-freedom adjustment parameters. The operation instructions are encrypted and sent to the corresponding construction equipment.

[0071] Compared with the prior art, the present invention has the following beneficial effects:

[0072] The bridge construction monitoring method and system based on the AI ​​platform provided in the present application photograph the bridge construction site from a first angle and a second angle respectively to obtain corresponding bridge construction images, and analyze the corresponding bridge construction images to obtain the existence status information of the first bridge component and the existence status information of the second bridge component, and perform multi-perspective positioning and identification of the bridge component; perform AI analysis on the existence status information of the first bridge component and the existence status information of the second bridge component to obtain the first posture information and the second posture information corresponding to the same component, so as to determine whether the same bridge component is in a normal posture state, thereby improving the accuracy of the detection of the posture state of the bridge component; when the same bridge component is not in a normal posture state, determine the operating parameters for adjusting the posture of the same bridge component, thereby generating operating instructions and sending them to the construction equipment, so as to facilitate the construction equipment to make timely and accurate posture adjustments to the bridge component, thereby reducing the positioning workload of the bridge component. BRIEF DESCRIPTION OF THE DRAWINGS

[0073] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the following briefly introduces the drawings required for use in the embodiments or the prior art descriptions. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without creative work. Among them:

[0074] Figure 1 A schematic flow chart of the bridge construction monitoring method based on the AI ​​platform provided by the present invention.

[0075] Figure 2 This is a schematic diagram of the structure of the bridge construction monitoring system based on the AI ​​platform provided by the present invention. DETAILED DESCRIPTION

[0076] In order to make the above-mentioned purposes, features and advantages of the present application more obvious and easy to understand, the specific implementation methods of the present application are described in detail below in conjunction with the accompanying drawings. It is to be understood that the specific embodiments described herein are only used to explain the present application, rather than to limit the present application. It should also be noted that, for ease of description, only some structures related to the present application are shown in the accompanying drawings, rather than all structures. Based on the embodiments in the present application, all other embodiments obtained by ordinary technicians in this field without making creative work are within the scope of protection of this application.

[0077] The terms "including" and "having" and any variations thereof in this application are intended to cover non-exclusive inclusions. For example, a process, method, system, product or device comprising a series of steps or units is not limited to the listed steps or units, but may optionally include steps or units not listed, or may optionally include other steps or units inherent to these processes, methods, products or devices.

[0078] Reference to "embodiments" herein means that a particular feature, structure, or characteristic described in conjunction with the embodiments may be included in at least one embodiment of the present application. The appearance of the phrase in various locations in the specification does not necessarily refer to the same embodiment, nor is it an independent or alternative embodiment that is mutually exclusive with other embodiments. It is explicitly and implicitly understood by those skilled in the art that the embodiments described herein may be combined with other embodiments.

[0079] See also Figure 1 As shown, a bridge construction monitoring method based on an AI platform provided in an embodiment of the present application includes:

[0080] Photographing the bridge construction site from a first angle to obtain a first bridge construction image; analyzing the first bridge construction image to obtain the existence status information of the first bridge component at the bridge construction site;

[0081] photographing the bridge construction site from a second angle to obtain a second bridge construction image; analyzing the second bridge construction image to obtain the second bridge component existence status information of the bridge construction site;

[0082] Uploading the first bridge component existence status information and the second bridge component existence status information to the AI ​​platform, extracting the first posture information and the second posture information corresponding to the same bridge component; judging whether the same bridge component is in a normal posture state based on the first posture information and the second posture information;

[0083] When the same bridge component is not in a normal posture state, the operating parameters for adjusting the posture of the same bridge component are determined based on the first bridge construction image and the second bridge construction image, and corresponding operating instructions are generated based on the operating parameters and sent to the corresponding construction equipment.

[0084] The beneficial effects of the above embodiments are as follows: the bridge construction monitoring method based on the AI ​​platform photographs the bridge construction site from a first angle and a second angle respectively to obtain corresponding bridge construction images, and analyzes the corresponding bridge construction images to obtain the existence status information of the first bridge component and the existence status information of the second bridge component, and performs multi-perspective positioning and identification of the bridge component; AI analysis is performed on the existence status information of the first bridge component and the second bridge component to obtain the first posture information and the second posture information corresponding to the same component, thereby judging whether the same bridge component is in a normal posture state, thereby improving the accuracy of bridge component posture state detection; when the same bridge component is not in a normal posture state, determining the operating parameters for adjusting the posture of the same bridge component, thereby generating operating instructions and sending them to the construction equipment, thereby facilitating the construction equipment to make timely and accurate posture adjustments to the bridge component, thereby reducing the workload of bridge component positioning.

[0085] In another embodiment, a bridge construction site is photographed from a first angle to obtain a first bridge construction image; and the first bridge construction image is analyzed to obtain the first bridge component existence status information of the bridge construction site, including:

[0086] Taking binocular photos of the bridge construction site from a ground perspective to obtain a binocular image of the bridge construction site; generating a first three-dimensional bridge construction image based on the binocular parallax of the binocular image;

[0087] The first bridge construction three-dimensional image is subjected to picture pixel contour recognition processing to obtain first edge contour distribution state information of all bridge components at the bridge construction site.

[0088] The beneficial effect of the above-mentioned embodiment is to perform binocular photography of the bridge construction site from a ground perspective, obtain a binocular image of the bridge construction site, and generate a corresponding first three-dimensional image of the bridge construction, so that the placement and distribution status of all bridge components inside the bridge construction site in the ground perspective observation scene can be comprehensively and accurately identified.

[0089] In another embodiment, the bridge construction site is photographed from a second angle to obtain a second bridge construction image; the second bridge construction image is analyzed to obtain the second bridge component existence status information of the bridge construction site, including:

[0090] Scan and photograph the bridge construction site from a high altitude to obtain a panoramic image of the bridge construction site; perform three-dimensional conversion on the panoramic image to obtain a second bridge construction three-dimensional image;

[0091] The second bridge construction three-dimensional image is subjected to picture pixel contour recognition processing to obtain second edge contour distribution state information of all bridge components at the bridge construction site.

[0092] The beneficial effect of the above-mentioned embodiment is that the bridge construction site is scanned and photographed from a high altitude perspective to obtain a panoramic image of the bridge construction site and generate a corresponding second bridge construction three-dimensional image, so that the placement and distribution status of all bridge components inside the bridge construction site in the high-altitude observation scene can be comprehensively and accurately identified.

[0093] In another embodiment, scanning and photographing the bridge construction site from a high altitude angle includes:

[0094] Extract the panoramic image corresponding to each scanning shot;

[0095] Comparing the similarity of the panoramic images corresponding to each two adjacent scanning moments to obtain the similarity value between the panoramic images corresponding to each two adjacent scanning moments;

[0096] Extract the scanning time interval between every two adjacent scanning moments;

[0097] Obtaining a scanning time interval adjustment coefficient by using the scanning time interval between each two adjacent scanning moments and a similarity value between panoramic images corresponding to each two adjacent scanning moments;

[0098] The scanning time interval adjustment coefficient is obtained by the following formula:

[0099]

[0100] Wherein, K represents the scanning time interval adjustment coefficient; n represents the number of panoramic image groups corresponding to every two adjacent scanning moments; S i S represents the similarity value of the panoramic images corresponding to two adjacent scanning moments of the i-th group; i+1 T represents the similarity value of the panoramic images corresponding to two adjacent scanning moments of the i+1th group; gi T represents the scanning time interval of the panoramic image corresponding to two adjacent scanning moments of the i-th group; gi+1 represents the scanning time interval of the panoramic image corresponding to two adjacent scanning moments of the i+1th group; S min represents the minimum similarity value of the panoramic images corresponding to n groups of two adjacent scanning moments; T gmin S represents the scanning time interval corresponding to the minimum similarity value of the panoramic images corresponding to n groups of two adjacent scanning moments; b T represents the standard deviation of the similarity values ​​of the panoramic images corresponding to n groups of two adjacent scanning moments; gb It represents the standard deviation of the scanning time interval corresponding to two adjacent scanning moments of n groups;

[0101] Comparing the scanning time interval adjustment coefficient with a preset adjustment coefficient threshold;

[0102] When the scanning time interval adjustment coefficient is lower than a preset adjustment coefficient threshold, the scanning time interval is not adjusted;

[0103] When the scanning time interval adjustment coefficient is not lower than a preset adjustment coefficient threshold, retrieving the scanning time interval adjustment coefficient;

[0104] Using the scanning time interval adjustment coefficient to adjust the scanning time interval to obtain an adjusted scanning time interval;

[0105] The adjusted scanning time interval is obtained by the following formula:

[0106]

[0107] Among them, T gt Represents the adjusted scanning time interval; T g represents the scanning time interval before adjustment; K represents the scanning time interval adjustment coefficient; K0 represents the preset adjustment coefficient threshold; S k S represents the similarity value between the corresponding panoramic image when the scanning time interval adjustment coefficient is not less than the preset adjustment coefficient threshold and the corresponding panoramic image taken in the previous scanning; p Represents the average value of similarity between n groups of panoramic images corresponding to two adjacent scanning moments;

[0108] The bridge construction site is scanned and photographed according to the adjusted scanning time interval.

[0109] The beneficial effect of the above embodiment is that by calculating the similarity value between the panoramic images of two adjacent scans, the scheme can intelligently judge the degree of change of the construction site. When the site changes little (i.e., the similarity value is high), the scanning time interval can be appropriately extended to reduce the number of unnecessary shots and improve the shooting efficiency. When the construction site changes significantly (i.e., the similarity value is low), the scheme will automatically shorten the scanning time interval to ensure that these changes can be captured, thereby maintaining the continuity and accuracy of the shooting. This is of great significance for monitoring the construction progress and timely discovering potential problems. The scheme introduces a scanning time interval adjustment coefficient K, which is dynamically calculated based on the similarity values ​​of adjacent scanned images and the scanning time interval. This adaptive mechanism enables the shooting strategy to be flexibly adjusted according to the actual situation of the construction site, improving the robustness and adaptability of the entire system. Through the preset adjustment coefficient threshold K0, the scheme can automatically determine whether the scanning time interval needs to be adjusted, and perform intelligent management accordingly. This reduces the need for manual intervention and improves the degree of automation of the shooting process. The scheme is based on panoramic images and similarity calculations, which are relatively mature in the field of image processing. Therefore, the scheme is relatively simple in implementation and maintenance, and is easy to apply and promote in actual projects.

[0110] At the same time, by dynamically adjusting the scanning time interval, the technical solution can significantly reduce the number of unnecessary scans, especially when the construction site does not change much. This directly improves scanning efficiency and reduces resource consumption, such as electricity, storage space, and processing power. When the construction site changes significantly, the solution can quickly shorten the scanning time interval to ensure that these changes are captured. This ensures the continuity and accuracy of the data, helps to promptly discover potential problems, and improves the level of refinement of construction management. The technical solution realizes dynamic adjustment of the scanning time interval by introducing the scanning time interval adjustment coefficient K. This adaptive mechanism enables the solution to be flexibly adjusted according to the actual needs in different scenarios, improving the flexibility and adaptability of the system. Through the preset adjustment coefficient threshold K0 and the automated adjustment process, the solution realizes intelligent management of the scanning time interval. This reduces the need for manual intervention, improves the degree of automation of the system, and reduces the difficulty and complexity of operation. The technical solution can cope with the uncertainty of changes in the construction site and maintain the stability and reliability of the system by dynamically adjusting the scanning strategy. This enhances the robustness of the system, enabling it to continue to play a role in a complex and changing construction environment. By reasonably adjusting the scanning time interval, the solution can optimize the utilization of resources. While maintaining data accuracy, it reduces the usage of storage and processing resources and improves resource utilization efficiency.

[0111] In summary, the technical effects of the above technical solutions in terms of performance indicators are mainly reflected in the improvement of scanning efficiency, data accuracy, adaptability and flexibility, intelligence level, system robustness and resource optimization. These effects jointly improve the level of monitoring and management of bridge construction sites, and help ensure the smooth progress and safety of construction projects. At the same time, the above technical solutions not only improve the shooting efficiency and ensure the shooting accuracy by dynamically adjusting the scanning time interval, but also enhance the adaptability and intelligent management level of the system. This is of great significance for the monitoring and management of bridge construction sites, and helps to improve the safety and efficiency of the entire project.

[0112] In another embodiment, the first bridge component existence status information and the second bridge component existence status information are uploaded to the AI ​​platform, and the first posture information and the second posture information corresponding to the same bridge component are extracted therefrom; based on the first posture information and the second posture information, it is determined whether the same bridge component is in a normal posture state, including:

[0113] The first edge contour distribution state information and the second edge contour distribution state information are uploaded to the AI ​​platform, and AI recognition is performed on the first edge contour distribution state information and the second edge contour distribution state information respectively to obtain the edge contour spatial distribution information of all bridge components in the first bridge construction three-dimensional image and the edge contour spatial distribution information of all bridge components in the second bridge construction three-dimensional image;

[0114] Compare the edge contour spatial distribution information of all bridge components in the first bridge construction image and the edge contour spatial distribution information of all bridge components in the second bridge construction image, and extract the first placement posture information and the second placement posture information corresponding to the same bridge component;

[0115] The first placement posture information and the second placement posture information are integrated into the same coordinate space for comparison to obtain corresponding posture deviation values; if the posture deviation value is less than a preset deviation threshold, it is determined that the same bridge component is in a normal posture state; otherwise, it is determined that the same bridge is not in a normal posture state.

[0116] The beneficial effect of the above embodiment is that the first edge contour distribution state information and the second edge contour distribution state information are uploaded to the AI ​​platform for AI recognition, so that the edge contour spatial distribution information of all bridge components inside the bridge construction site can be identified from two different shooting field directions, so as to extract the first placement posture information and the second placement posture information corresponding to the same bridge component, and realize the multi-faceted placement posture identification of the same bridge component. Then the first placement posture information and the second placement posture information are integrated into the same coordinate space for comparison to obtain the corresponding posture deviation value, so as to quantitatively determine whether the same bridge component is in a normal posture state.

[0117] In another embodiment, when the same bridge component is not in a normal posture state, an operation parameter for adjusting the posture of the same bridge component is determined based on the first bridge construction image and the second bridge construction image, and a corresponding operation instruction is generated based on the operation parameter and sent to the corresponding construction equipment, including:

[0118] When the same bridge component is not in a normal posture state, the first bridge construction three-dimensional image and the second bridge construction three-dimensional image are analyzed to determine the six-degree-of-freedom adjustment parameters for adjusting the posture of the same bridge component, and corresponding operation instructions are generated based on the six-degree-of-freedom adjustment parameters. The operation instructions are encrypted and sent to the corresponding construction equipment.

[0119] The beneficial effect of the above embodiment is that when the same bridge component is not in a normal posture state, the first bridge construction three-dimensional image and the second bridge construction three-dimensional image are analyzed to determine the six-degree-of-freedom adjustment parameters for adjusting the posture of the same bridge component. The six-degree-of-freedom adjustment parameters may include the displacement in the three-axis directions of the three-dimensional space and the yaw angle, pitch angle, and roll angle. Based on the six-degree-of-freedom adjustment parameters, corresponding operation instructions are generated, and the operation instructions are encrypted and sent to the corresponding construction equipment, so that construction equipment such as elevators or cranes can accurately adjust the bridge component.

[0120] See also Figure 2 As shown, the bridge construction monitoring system based on the AI ​​platform provided in one embodiment of the present application includes:

[0121] A first image shooting and analysis module is used to shoot the bridge construction site from a first angle to obtain a first bridge construction image; analyze the first bridge construction image to obtain the existence status information of the first bridge component of the bridge construction site;

[0122] The second image shooting and analysis module is used to shoot the bridge construction site from a second angle to obtain a second bridge construction image; analyze the second bridge construction image to obtain the second bridge component existence status information of the bridge construction site;

[0123] An AI processing module is used to upload the existence status information of the first bridge component and the existence status information of the second bridge component to the AI ​​platform, and extract the first posture information and the second posture information corresponding to the same bridge component; based on the first posture information and the second posture information, determine whether the same bridge component is in a normal posture state;

[0124] The bridge component adjustment module is used to determine the operating parameters for adjusting the posture of the same bridge component based on the first bridge construction image and the second bridge construction image when the same bridge component is not in a normal posture state, and generate corresponding operating instructions based on the operating parameters, and send them to the corresponding construction equipment.

[0125] The beneficial effects of the above embodiments are as follows: the bridge construction monitoring system based on the AI ​​platform photographs the bridge construction site from a first angle and a second angle respectively to obtain corresponding bridge construction images, and analyzes the corresponding bridge construction images to obtain the existence status information of the first bridge component and the existence status information of the second bridge component, and performs multi-perspective positioning and identification of the bridge component; performs AI analysis on the existence status information of the first bridge component and the second bridge component to obtain the first posture information and the second posture information corresponding to the same component, thereby judging whether the same bridge component is in a normal posture state, and improving the accuracy of the detection of the posture state of the bridge component; when the same bridge component is not in a normal posture state, determining the operating parameters for adjusting the posture of the same bridge component, thereby generating operating instructions and sending them to the construction equipment, so as to facilitate the construction equipment to make timely and accurate posture adjustments to the bridge component, and reduce the positioning workload of the bridge component.

[0126] In another embodiment, the first image shooting and analysis module is used to shoot the bridge construction site from a first angle to obtain a first bridge construction image; analyze the first bridge construction image to obtain the first bridge component existence status information of the bridge construction site, including:

[0127] Taking binocular photos of the bridge construction site from a ground perspective to obtain a binocular image of the bridge construction site; generating a first three-dimensional bridge construction image based on the binocular parallax of the binocular image;

[0128] The first bridge construction three-dimensional image is subjected to picture pixel contour recognition processing to obtain first edge contour distribution state information of all bridge components at the bridge construction site.

[0129] The beneficial effect of the above-mentioned embodiment is to perform binocular photography of the bridge construction site from a ground perspective, obtain a binocular image of the bridge construction site, and generate a corresponding first three-dimensional image of the bridge construction, so that the placement and distribution status of all bridge components inside the bridge construction site in the ground perspective observation scene can be comprehensively and accurately identified.

[0130] In another embodiment, the second image shooting and analysis module is used to shoot the bridge construction site from a second angle to obtain a second bridge construction image; analyze the second bridge construction image to obtain the second bridge component existence status information of the bridge construction site, including:

[0131] Scan and photograph the bridge construction site from a high altitude to obtain a panoramic image of the bridge construction site; perform three-dimensional conversion on the panoramic image to obtain a second bridge construction three-dimensional image;

[0132] The second bridge construction three-dimensional image is subjected to picture pixel contour recognition processing to obtain second edge contour distribution state information of all bridge components at the bridge construction site.

[0133] The beneficial effect of the above-mentioned embodiment is that the bridge construction site is scanned and photographed from a high altitude perspective to obtain a panoramic image of the bridge construction site and generate a corresponding second bridge construction three-dimensional image, so that the placement and distribution status of all bridge components inside the bridge construction site in the high-altitude observation scene can be comprehensively and accurately identified.

[0134] In another embodiment, scanning and photographing the bridge construction site from a high altitude angle includes:

[0135] Extract the panoramic image corresponding to each scanning shot;

[0136] Comparing the similarity of the panoramic images corresponding to each two adjacent scanning moments to obtain the similarity value between the panoramic images corresponding to each two adjacent scanning moments;

[0137] Extract the scanning time interval between every two adjacent scanning moments;

[0138] Obtaining a scanning time interval adjustment coefficient by using the scanning time interval between each two adjacent scanning moments and a similarity value between panoramic images corresponding to each two adjacent scanning moments;

[0139] The scanning time interval adjustment coefficient is obtained by the following formula:

[0140]

[0141] Wherein, K represents the scanning time interval adjustment coefficient; n represents the number of panoramic image groups corresponding to every two adjacent scanning moments; Si S represents the similarity value of the panoramic images corresponding to two adjacent scanning moments of the i-th group; i+1 T represents the similarity value of the panoramic images corresponding to two adjacent scanning moments of the i+1th group; gi T represents the scanning time interval of the panoramic image corresponding to two adjacent scanning moments of the i-th group; gi+1 represents the scanning time interval of the panoramic image corresponding to two adjacent scanning moments of the i+1th group; S min represents the minimum similarity value of the panoramic images corresponding to n groups of two adjacent scanning moments; T gmin S represents the scanning time interval corresponding to the minimum similarity value of the panoramic images corresponding to n groups of two adjacent scanning moments; b T represents the standard deviation of the similarity values ​​of the panoramic images corresponding to n groups of two adjacent scanning moments; gb It represents the standard deviation of the scanning time interval corresponding to two adjacent scanning moments of n groups;

[0142] Comparing the scanning time interval adjustment coefficient with a preset adjustment coefficient threshold;

[0143] When the scanning time interval adjustment coefficient is lower than a preset adjustment coefficient threshold, the scanning time interval is not adjusted;

[0144] When the scanning time interval adjustment coefficient is not lower than a preset adjustment coefficient threshold, retrieving the scanning time interval adjustment coefficient;

[0145] Using the scanning time interval adjustment coefficient to adjust the scanning time interval to obtain an adjusted scanning time interval;

[0146] The adjusted scanning time interval is obtained by the following formula:

[0147]

[0148] Among them, T gt Represents the adjusted scanning time interval; T g represents the scanning time interval before adjustment; K represents the scanning time interval adjustment coefficient; K0 represents the preset adjustment coefficient threshold; S k S represents the similarity value between the corresponding panoramic image when the scanning time interval adjustment coefficient is not less than the preset adjustment coefficient threshold and the corresponding panoramic image taken in the previous scanning; p Represents the average value of similarity between n groups of panoramic images corresponding to two adjacent scanning moments;

[0149] The bridge construction site is scanned and photographed according to the adjusted scanning time interval.

[0150] The beneficial effect of the above embodiment is that by calculating the similarity value between the panoramic images of two adjacent scans, the scheme can intelligently judge the degree of change of the construction site. When the site changes little (i.e., the similarity value is high), the scanning time interval can be appropriately extended to reduce the number of unnecessary shots and improve the shooting efficiency. When the construction site changes significantly (i.e., the similarity value is low), the scheme will automatically shorten the scanning time interval to ensure that these changes can be captured, thereby maintaining the continuity and accuracy of the shooting. This is of great significance for monitoring the construction progress and timely discovering potential problems. The scheme introduces a scanning time interval adjustment coefficient K, which is dynamically calculated based on the similarity values ​​of adjacent scanned images and the scanning time interval. This adaptive mechanism enables the shooting strategy to be flexibly adjusted according to the actual situation of the construction site, improving the robustness and adaptability of the entire system. Through the preset adjustment coefficient threshold K0, the scheme can automatically determine whether the scanning time interval needs to be adjusted, and perform intelligent management accordingly. This reduces the need for manual intervention and improves the degree of automation of the shooting process. The scheme is based on panoramic images and similarity calculations, which are relatively mature in the field of image processing. Therefore, the scheme is relatively simple in implementation and maintenance, and is easy to apply and promote in actual projects.

[0151] At the same time, by dynamically adjusting the scanning time interval, the technical solution can significantly reduce the number of unnecessary scans, especially when the construction site does not change much. This directly improves scanning efficiency and reduces resource consumption, such as electricity, storage space, and processing power. When the construction site changes significantly, the solution can quickly shorten the scanning time interval to ensure that these changes are captured. This ensures the continuity and accuracy of the data, helps to promptly discover potential problems, and improves the level of refinement of construction management. The technical solution realizes dynamic adjustment of the scanning time interval by introducing the scanning time interval adjustment coefficient K. This adaptive mechanism enables the solution to be flexibly adjusted according to the actual needs in different scenarios, improving the flexibility and adaptability of the system. Through the preset adjustment coefficient threshold K0 and the automated adjustment process, the solution realizes intelligent management of the scanning time interval. This reduces the need for manual intervention, improves the degree of automation of the system, and reduces the difficulty and complexity of operation. The technical solution can cope with the uncertainty of changes in the construction site and maintain the stability and reliability of the system by dynamically adjusting the scanning strategy. This enhances the robustness of the system, enabling it to continue to play a role in a complex and changing construction environment. By reasonably adjusting the scanning time interval, the solution can optimize the utilization of resources. While maintaining data accuracy, it reduces the usage of storage and processing resources and improves resource utilization efficiency.

[0152] In summary, the technical effects of the above technical solutions in terms of performance indicators are mainly reflected in the improvement of scanning efficiency, data accuracy, adaptability and flexibility, intelligence level, system robustness and resource optimization. These effects jointly improve the level of monitoring and management of bridge construction sites, and help ensure the smooth progress and safety of construction projects. At the same time, the above technical solutions not only improve the shooting efficiency and ensure the shooting accuracy by dynamically adjusting the scanning time interval, but also enhance the adaptability and intelligent management level of the system. This is of great significance for the monitoring and management of bridge construction sites, and helps to improve the safety and efficiency of the entire project.

[0153] In another embodiment, the AI ​​processing module is used to upload the first bridge component existence status information and the second bridge component existence status information to the AI ​​platform, and extract the first posture information and the second posture information corresponding to the same bridge component; based on the first posture information and the second posture information, determine whether the same bridge component is in a normal posture state, including:

[0154] The first edge contour distribution state information and the second edge contour distribution state information are uploaded to the AI ​​platform, and AI recognition is performed on the first edge contour distribution state information and the second edge contour distribution state information respectively to obtain the edge contour spatial distribution information of all bridge components in the first bridge construction three-dimensional image and the edge contour spatial distribution information of all bridge components in the second bridge construction three-dimensional image;

[0155] Compare the edge contour spatial distribution information of all bridge components in the first bridge construction image and the edge contour spatial distribution information of all bridge components in the second bridge construction image, and extract the first placement posture information and the second placement posture information corresponding to the same bridge component;

[0156] The first placement posture information and the second placement posture information are integrated into the same coordinate space for comparison to obtain corresponding posture deviation values; if the posture deviation value is less than a preset deviation threshold, it is determined that the same bridge component is in a normal posture state; otherwise, it is determined that the same bridge is not in a normal posture state.

[0157] The beneficial effect of the above embodiment is that the first edge contour distribution state information and the second edge contour distribution state information are uploaded to the AI ​​platform for AI recognition, so that the edge contour spatial distribution information of all bridge components inside the bridge construction site can be identified from two different shooting field directions, so as to extract the first placement posture information and the second placement posture information corresponding to the same bridge component, and realize the multi-faceted placement posture identification of the same bridge component. Then the first placement posture information and the second placement posture information are integrated into the same coordinate space for comparison to obtain the corresponding posture deviation value, so as to quantitatively determine whether the same bridge component is in a normal posture state.

[0158] In another embodiment, the bridge component adjustment module is used to determine, when the same bridge component is not in a normal posture state, based on the first bridge construction image and the second bridge construction image, an operation parameter for adjusting the posture of the same bridge component, and generate a corresponding operation instruction based on the operation parameter, and send it to the corresponding construction equipment, including:

[0159] When the same bridge component is not in a normal posture state, the first bridge construction three-dimensional image and the second bridge construction three-dimensional image are analyzed to determine the six-degree-of-freedom adjustment parameters for adjusting the posture of the same bridge component, and corresponding operation instructions are generated based on the six-degree-of-freedom adjustment parameters. The operation instructions are encrypted and sent to the corresponding construction equipment.

[0160] The beneficial effect of the above embodiment is that when the same bridge component is not in a normal posture state, the first bridge construction three-dimensional image and the second bridge construction three-dimensional image are analyzed to determine the six-degree-of-freedom adjustment parameters for adjusting the posture of the same bridge component. The six-degree-of-freedom adjustment parameters may include the displacement in the three-axis directions of the three-dimensional space and the yaw angle, pitch angle, and roll angle. Based on the six-degree-of-freedom adjustment parameters, corresponding operation instructions are generated, and the operation instructions are encrypted and sent to the corresponding construction equipment, so that construction equipment such as elevators or cranes can accurately adjust the bridge component.

[0161] In general, the bridge construction monitoring method and system based on the AI ​​platform photographs the bridge construction site from a first angle and a second angle respectively to obtain corresponding bridge construction images, and analyzes the corresponding bridge construction images to obtain the existence status information of the first bridge component and the existence status information of the second bridge component, and performs multi-perspective positioning and identification of the bridge component; AI analysis is performed on the existence status information of the first bridge component and the second bridge component to obtain the first posture information and the second posture information corresponding to the same component, so as to determine whether the same bridge component is in a normal posture state, thereby improving the accuracy of the detection of the posture state of the bridge component; when the same bridge component is not in a normal posture state, the operating parameters for adjusting the posture of the same bridge component are determined, so as to generate operating instructions and send them to the construction equipment, so as to facilitate the construction equipment to make timely and accurate posture adjustments to the bridge component, thereby reducing the positioning workload of the bridge component.

[0162] The above is only a specific implementation of the present invention, and any other improvements made based on the concept of the present invention are considered to be within the protection scope of the present invention.

Claims

1. A bridge construction monitoring method based on an AI platform, characterized in that: include: Photographing the bridge construction site from a first angle to obtain a first bridge construction image; analyzing the first bridge construction image to obtain the existence status information of the first bridge component at the bridge construction site; photographing the bridge construction site from a second angle to obtain a second bridge construction image; analyzing the second bridge construction image to obtain the second bridge component existence status information of the bridge construction site; Uploading the first bridge component existence status information and the second bridge component existence status information to the AI ​​platform, and extracting the first posture information and the second posture information corresponding to the same bridge component; Based on the first posture information and the second posture information, determining whether the same bridge component is in a normal posture state; When the same bridge component is not in a normal posture state, the operating parameters for adjusting the posture of the same bridge component are determined based on the first bridge construction image and the second bridge construction image, and corresponding operating instructions are generated based on the operating parameters and sent to the corresponding construction equipment.

2. The bridge construction monitoring method based on the AI ​​platform according to claim 1, characterized in that: The bridge construction site is photographed from a first angle to obtain a first bridge construction image; the first bridge construction image is analyzed to obtain the first bridge component existence status information of the bridge construction site, including: A binocular photograph is taken of a bridge construction site from a ground perspective to obtain a binocular image of the bridge construction site; a first three-dimensional bridge construction image is generated based on the binocular parallax of the binocular image; and pixel contour recognition processing is performed on the first three-dimensional bridge construction image to obtain first edge contour distribution status information of all bridge components at the bridge construction site.

3. The bridge construction monitoring method based on the AI ​​platform as claimed in claim 2, characterized in that: The bridge construction site is photographed from a second angle to obtain a second bridge construction image; the second bridge construction image is analyzed to obtain the second bridge component existence status information of the bridge construction site, including: Scan and photograph the bridge construction site from a high altitude to obtain a panoramic image of the bridge construction site; perform three-dimensional conversion on the panoramic image to obtain a second three-dimensional image of the bridge construction site; The second bridge construction three-dimensional image is subjected to picture pixel contour recognition processing to obtain second edge contour distribution state information of all bridge components at the bridge construction site.

4. The bridge construction monitoring method based on the AI ​​platform as claimed in claim 3, characterized in that: Uploading the first bridge component existence status information and the second bridge component existence status information to the AI ​​platform, and extracting the first posture information and the second posture information corresponding to the same bridge component; Based on the first posture information and the second posture information, determining whether the same bridge component is in a normal posture state includes: Uploading the first edge contour distribution state information and the second edge contour distribution state information to the AI ​​platform, performing AI recognition on the first edge contour distribution state information and the second edge contour distribution state information, respectively, to obtain the edge contour spatial distribution information of all bridge components in the first bridge construction 3D image and the edge contour spatial distribution information of all bridge components in the second bridge construction 3D image; Compare the edge contour spatial distribution information of all bridge components in the first bridge construction image and the edge contour spatial distribution information of all bridge components in the second bridge construction image, and extract the first placement posture information and the second placement posture information corresponding to the same bridge component; The first placement posture information and the second placement posture information are integrated into the same coordinate space for comparison to obtain corresponding posture deviation values; if the posture deviation value is less than a preset deviation threshold, it is judged that the same bridge component is in a normal posture state; otherwise, it is judged that the same bridge component is not in a normal posture state.

5. The bridge construction monitoring method based on the AI ​​platform according to claim 4, characterized in that: When the same bridge component is not in a normal posture state, based on the first bridge construction image and the second bridge construction image, determining operation parameters for adjusting the posture of the same bridge component, and generating corresponding operation instructions based on the operation parameters, and sending them to corresponding construction equipment, including: When the same bridge component is not in a normal posture state, the first bridge construction three-dimensional image and the second bridge construction three-dimensional image are analyzed to determine the six-degree-of-freedom adjustment parameters for adjusting the posture of the same bridge component, and corresponding operation instructions are generated based on the six-degree-of-freedom adjustment parameters. The operation instructions are encrypted and sent to the corresponding construction equipment.

6. The bridge construction monitoring system based on AI platform is characterized by: include: A first image shooting and analysis module is used to shoot the bridge construction site from a first angle to obtain a first bridge construction image; analyze the first bridge construction image to obtain the existence status information of the first bridge component of the bridge construction site; A second image shooting and analysis module is used to shoot the bridge construction site from a second angle to obtain a second bridge construction image; analyze the second bridge construction image to obtain the second bridge component existence status information of the bridge construction site; An AI processing module is used to upload the first bridge component existence state information and the second bridge component existence state information to an AI platform, and extract the first posture information and the second posture information corresponding to the same bridge component; Based on the first posture information and the second posture information, determining whether the same bridge component is in a normal posture state; The bridge component adjustment module is used to determine the operating parameters for adjusting the posture of the same bridge component based on the first bridge construction image and the second bridge construction image when the same bridge component is not in a normal posture state, and generate corresponding operating instructions based on the operating parameters, and send them to the corresponding construction equipment.

7. The bridge construction monitoring system based on the AI ​​platform as claimed in claim 6, characterized in that: The first image shooting and analysis module is used to shoot the bridge construction site from a first angle to obtain a first bridge construction image; analyze the first bridge construction image to obtain the first bridge component existence status information of the bridge construction site, including: A binocular photograph is taken of a bridge construction site from a ground perspective to obtain a binocular image of the bridge construction site; a first three-dimensional bridge construction image is generated based on the binocular parallax of the binocular image; and pixel contour recognition processing is performed on the first three-dimensional bridge construction image to obtain first edge contour distribution status information of all bridge components at the bridge construction site.

8. The bridge construction monitoring system based on the AI ​​platform as claimed in claim 7, characterized in that: The second image shooting and analysis module is used to shoot the bridge construction site from a second angle to obtain a second bridge construction image; analyze the second bridge construction image to obtain the second bridge component existence status information of the bridge construction site, including: Scan and photograph the bridge construction site from a high altitude to obtain a panoramic image of the bridge construction site; perform three-dimensional conversion on the panoramic image to obtain a second three-dimensional image of the bridge construction site; The second bridge construction three-dimensional image is subjected to picture pixel contour recognition processing to obtain second edge contour distribution state information of all bridge components at the bridge construction site.

9. The bridge construction monitoring system based on the AI ​​platform as claimed in claim 8, characterized in that: The AI ​​processing module is used to upload the first bridge component existence status information and the second bridge component existence status information to the AI ​​platform, and extract the first posture information and the second posture information corresponding to the same bridge component; Based on the first posture information and the second posture information, determining whether the same bridge component is in a normal posture state includes: Uploading the first edge contour distribution state information and the second edge contour distribution state information to the AI ​​platform, performing AI recognition on the first edge contour distribution state information and the second edge contour distribution state information, respectively, to obtain the edge contour spatial distribution information of all bridge components in the first bridge construction 3D image and the edge contour spatial distribution information of all bridge components in the second bridge construction 3D image; Compare the edge contour spatial distribution information of all bridge components in the first bridge construction image and the edge contour spatial distribution information of all bridge components in the second bridge construction image, and extract the first placement posture information and the second placement posture information corresponding to the same bridge component; The first placement posture information and the second placement posture information are integrated into the same coordinate space for comparison to obtain corresponding posture deviation values; if the posture deviation value is less than a preset deviation threshold, it is judged that the same bridge component is in a normal posture state; otherwise, it is judged that the same bridge component is not in a normal posture state.

10. The bridge construction monitoring system based on the AI ​​platform according to claim 9, characterized in that: The bridge component adjustment module is used to determine the operation parameters for adjusting the posture of the same bridge component based on the first bridge construction image and the second bridge construction image when the same bridge component is not in a normal posture state, and generate corresponding operation instructions based on the operation parameters, and send them to corresponding construction equipment, including: When the same bridge component is not in a normal posture state, the first bridge construction three-dimensional image and the second bridge construction three-dimensional image are analyzed to determine the six-degree-of-freedom adjustment parameters for adjusting the posture of the same bridge component, and corresponding operation instructions are generated based on the six-degree-of-freedom adjustment parameters. The operation instructions are encrypted and sent to the corresponding construction equipment.

Images