Cross-key parameter recognition method for cast-in-situ box girder images based on target detection and key information extraction

Abstract

The present invention relates to a method for identifying key parameters across cast-in-situ box girder images based on target detection and key information extraction, comprising the following steps: obtaining and preprocessing bridge design drawings; obtaining and training a bridge structure detection model; intercepting bridge structure images under different views based on images after target detection of the multi-view structure of the bridge; selecting different key parameters to be identified; obtaining an editable Excel document; and summarizing the key parameter information of the bridge structure within the same span into a new Excel document. The present invention simultaneously assigns semantic information to key parameters in the drawings, allowing parameter names and values ​​to correspond to each other, thereby achieving modularization of parameter identification for each plan view and being able to efficiently and automatically extract key parameters from PDF drawings. Finally, a method for identifying and integrating by span is adopted to successfully solve the parameter identification problem of multiple views, providing a beneficial exploration for artificial intelligence algorithms to understand engineering design drawings. The method can quickly and accurately extract key parameters from multiple views in PDF drawings of cast-in-situ box girders.

Description

Technical Field

[0001] The present invention relates to the technical field of PDF image recognition, and in particular to a method for identifying cross-key parameters of cast-in-situ box girder images based on target detection and key information extraction. Background Art

[0002] Bridge design drawings carry core information about bridge engineering, including dimensions, locations, material specifications, and key construction requirements. This information is crucial for ensuring the structural safety, constructibility, and overall quality of bridges. Effectively identifying and archiving design drawings not only provides crucial data for 3D reconstruction but also helps construction workers better understand bridge structures, improve installation and assembly efficiency, and reduce errors and delays during construction. In practical applications, parameter recognition in bridge drawings presents numerous challenges. First, different designers have varying design habits, leading to variations in annotations and symbols on drawings, which complicates recognition. Second, parameters on PDF drawings of bridge structures are typically simply numbers, lacking semantic information, making it difficult to understand their specific meaning. Furthermore, PDF design drawings of cast-in-place box girders are more complex, with multiple layers of information, symbol occlusion, and overlapping lines, hindering parameter recognition. Compared to architectural drawing recognition, bridge drawing recognition lacks publicly available datasets, requiring more parameters to be recognized and requiring higher accuracy. Finally, the parameters to be recognized in cast-in-place box girders are distributed across different plan views, requiring comprehensive recognition and parameter integration across multiple views.

[0003] With the rapid development of computer science, more and more researchers are devoting themselves to the study of drawing recognition technology, and related processing techniques are gradually maturing. Although mainstream CAD software now has professional drawing information management functions, it is still difficult to effectively extract key information from non-vectorized drawings, such as scanned documents. Current methods are only suitable for information extraction from vectorized drawings, or methods that use image feature matching suffer from poor robustness.

[0004] How to design a structural detection and key parameter identification method for the general structural drawing PDF design drawings of cast-in-place box girders has become a technical problem that needs to be solved urgently. Summary of the Invention

[0005] In order to solve the problem of difficulty in multi-view parameter recognition in PDF design drawings of general structural drawings of cast-in-situ box girders, the purpose of the present invention is to provide a cross-key parameter recognition method for cast-in-situ box girder images based on target detection and key information extraction, which can quickly and accurately extract key parameters of multiple views in PDF drawings of cast-in-situ box girders.

[0006] To achieve the above objectives, the present invention adopts the following technical solution: a method for identifying cross-key parameters of cast-in-situ box girder images based on target detection and key information extraction, the method comprising the following steps in order:

[0007] (1) Obtaining bridge design drawings and preprocessing them to obtain preprocessed bridge design drawings;

[0008] (2) Improve the YOLOv10 model to obtain a bridge structure detection model;

[0009] (3) Inputting the preprocessed bridge design drawings into the bridge structure detection model for training to obtain a trained bridge structure detection model;

[0010] (4) Obtain the design drawings of the bridge to be inspected and preprocess them, input the preprocessed design drawings of the bridge to be inspected into the trained bridge structure detection model, and the trained bridge structure detection model outputs an image after the multi-view structure target detection of the bridge, and the relative position of the bridge structure target is marked on the image after the multi-view structure target detection of the bridge;

[0011] (5) Based on the image of the bridge after multi-view structural target detection, intercept the bridge structure image under different views, wherein the different views include a cross-sectional view, an elevation view, a top plan view, and a bottom plan view;

[0012] (6) Select different key parameters to be identified based on the bridge structure images under different views;

[0013] (7) Input the bridge structure images under different views and the different key parameters to be identified into the paddleocr model for feature extraction and feature fusion to obtain multimodal features. The multimodal features are input into the vi_layoutxlm_ser model for prediction to obtain an editable Excel document.

[0014] (8) Read the editable Excel document, and merge the editable Excel document of the key parameters of each view into a new Excel document of the key parameter information of the bridge structure in the same span according to the logical relationship between the parameters of multiple views of the bridge structure in the same span.

[0015] The step (1) specifically includes the following steps in order:

[0016] (1a) Use the pdf2image open source library to convert archived PDF bridge design drawings into JPG images. The resolution of each JPG image is 3573 × 2526.

[0017] (1b) Perform data augmentation by rotating all JPG images and adding Gaussian noise;

[0018] (1c) Labeling: Use the Labelimg tool to label the bridge structure in multiple views of the data augmented image, which are divided into cross-section, elevation, top plan, and bottom plan. The labeling format is YOLO format.

[0019] The step (2) specifically refers to: improving the YOLOv10 model, replacing all C2f modules in the YOLOv10 model with the DualConv module; and replacing the detection head in the YOLOv10 model with the detection head of the RT-DETR model.

[0020] The step (5) specifically includes the following steps in sequence:

[0021] (5a) Obtain the target frame information in the image after the bridge multi-view structure target detection, that is, the relative position of the bridge structure target in the image. The target frame information includes:<class_label><x_center><y_center> <width>and <height>;<class_label> Refers to the target category label, which uses an integer to represent the category number to which the detected target belongs;<x_center> and<y_center> Refers to the x-coordinate and y-coordinate of the center point of the target frame respectively; <width>and <height>Refers to the width and height of the target box respectively;

[0022] (5b) Convert the target frame information into coordinate information according to the following conversion relationship:

[0023] left = int((x_center - width / 2) * image.shape[1])

[0024] top = int((y_center - height / 2) * image.shape[0])

[0025] right = int((x_center + width / 2) * image.shape[1])

[0026] bottom = int((y_center + height / 2) * image.shape[0])

[0027] In the formula, image.shape[1] represents the width of the image, image.shape[0] represents the height of the image, left represents the horizontal coordinate position of the left boundary of the target box in the image; right represents the horizontal coordinate position of the right boundary of the target box in the image; top represents the vertical coordinate position of the upper boundary of the target box in the image; bottom represents the vertical coordinate position of the lower boundary of the target box in the image;

[0028] For each target, a Python script was used to capture the detected bridge structure image after multi-view target detection based on the target's coordinate information. The bridge structure images under different views, namely, cross-section, elevation, top plan, and bottom plan, were captured. The captured bridge structure images under different views were named "original image name + view name" and saved in the corresponding folders for each view.

[0029] The step (6) specifically refers to:

[0030] The key parameters that need to be identified to determine the cross-section include: serial number, box beam upper flange width, box beam lower flange width, flange left cantilever width, flange right cantilever width, beam height, cantilever end thickness and cantilever root thickness, a total of 8 parameters;

[0031] The key parameters that need to be identified in the elevation drawing include: section symbol, pier number, span and beam height, a total of 4 parameters;

[0032] The key parameters that need to be identified to determine the top plan include: pier number, span and beam height, a total of 3 parameters;

[0033] The key parameters that need to be identified to determine the bottom plan include: pier number, span and beam height, a total of 3 parameters.

[0034] The step (7) specifically includes the following steps in order:

[0035] (7a) Annotating the bridge structure images under different views, marking the key parameters on the bridge structure images under different views, and obtaining the annotated bridge structure images under different views;

[0036] (7b) The annotated bridge structure images from different views are input into the PaddleOCR model for feature extraction, extracting text information, text coordinate information, and layout feature information. The extracted text information, text coordinate information, and layout feature information are then encoded and converted and fused into multimodal features. Subsequently, these multimodal features are input into the vi_layoutxlm_ser model for forward reasoning calculation to predict the category, location, and specific text content of each semantic entity in the image, and output the recognition result image and text file.

[0037] (7c) Processing the text file: The text file includes the image path and each text recognition result entry, which includes the transcription, bbox, points, pred_id and pred fields and their corresponding values; transcription represents the text content recognized in the image; bbox represents the bounding box coordinate information of the area where the recognized text is located, which is given in the form of a list, representing the upper left corner horizontal coordinate, upper left corner vertical coordinate, lower right corner horizontal coordinate, and lower right corner vertical coordinate in order; points is used to describe the outline coordinate information of the area where the text is located, which is presented in the form of polygon vertex coordinates, and each element is a list containing horizontal and vertical coordinates; pred_id represents the number of the prediction result, and different numbers correspond to different recognition categories; pred represents the key parameter name corresponding to the recognized text, and the unlabeled text category is "NONE";

[0038] Use the Python script to filter out all entries whose pred value is not "NONE" and the "transcription" entry in the dictionary where the entry is located, and generate an editable Excel document.

[0039] The step (8) specifically includes the following steps in order:

[0040] (8a) The logical relationship between the parameters of multiple views of the same span bridge structure includes:

[0041] The elevation, top plan and bottom plan are identified by the pier number to which they belong. Piers with the same pier number belong to the same span.

[0042] The relationship between the serial number of the cross-section drawing and the cut symbol number in the elevation drawing needs to be analyzed; within the same span, the serial number of the cross-section drawing must fall within the cut symbol number range between two pier numbers in the elevation drawing;

[0043] (8b) Based on the logical relationship between the parameters of multiple views of the bridge structure within the same span, the editable Excel documents are merged and summarized into a new Excel document containing the key parameter information of the bridge structure within the same span to realize parameter extraction.

[0044] It can be seen from the above technical solution that the beneficial effects of the present invention are: First, compared with the prior art, the present invention uses the improved YOLOv10 model to locate and intercept the image after multi-view structure target detection of the bridge, and intercepts the bridge structure images under different views, namely, cross-section, elevation, top plan and bottom plan; uses the paddleocr model and the vi_layoutxlm_ser model to extract the position and value of the key parameters that need to be identified in each view; filters the content in the text file of the recognition result according to specific conditions, and finally integrates the Excel document with the same cross-key parameters according to the logical relationship of the parameters between multiple views; second, the present invention replaces the original structure corresponding to the YOLOv10 model with the DualConv module and the RT-DETR detection head, which can not only reduce parameters but also improve the model's recognition of the features of components of different scales in the bridge design drawing. Extraction accuracy can also be combined with the real-time detection capability of the YOLOv10 model and the deep feature understanding capability of the RT-DETR detection head; thirdly, the present invention also assigns semantic information to the key parameters in the drawings, so that the parameter names and values ​​correspond to each other, realizes the modularization of the parameter identification of each plane drawing, and can efficiently and automatically extract the key parameters of the PDF drawings; finally, the method of identification and integration according to span is adopted to successfully solve the parameter identification problem of multiple views, and conducts useful exploration for the artificial intelligence algorithm to understand engineering design drawings; fourth, experimental results show that the accuracy and recall rate of the improved YOLOv10 model on the multi-view test set are 94.2% and 95.6% respectively; the key parameter results identified by the key information extraction method are basically consistent with the manual identification results; the present invention can quickly and accurately extract the key parameters of multiple views in the PDF drawings of cast-in-place box girders. BRIEF DESCRIPTION OF THE DRAWINGS

[0045] Figure 1 is a flow chart of the method of the present invention;

[0046] Figure 2 This is a schematic diagram of the key parameters to be identified for the cross-section of the cast-in-situ box girder;

[0047] Figure 3 This is a schematic diagram of the key parameters to be identified for the cast-in-situ box girder elevation;

[0048] Figure 4 This is a schematic diagram of the key parameters to be identified in the top plan of the cast-in-situ box girder;

[0049] Figure 5 This is a schematic diagram of the key parameters to be identified in the bottom plan of the cast-in-situ box girder;

[0050] Figure 6 This is the result diagram of cross-section identification of cast-in-place box girder;

[0051] Figure 7 This is the result diagram of cast-in-place box girder elevation identification;

[0052] Figure 8 This is the result diagram of the top plan of the cast-in-place box girder;

[0053] Figure 9 This is the result diagram of the bottom plane identification of the cast-in-place box girder;

[0054] Figure 10 This is a schematic diagram of the overall results of the key parameter extraction of the same span of the cast-in-situ box girder involved in the present invention. DETAILED DESCRIPTION

[0055] like Figure 1 As shown, a method for identifying cross-key parameters of cast-in-situ box girder images based on target detection and key information extraction includes the following steps in sequence:

[0056] (1) Obtaining bridge design drawings and preprocessing them to obtain preprocessed bridge design drawings;

[0057] (2) Improve the YOLOv10 model to obtain a bridge structure detection model;

[0058] (3) Inputting the preprocessed bridge design drawings into the bridge structure detection model for training to obtain a trained bridge structure detection model;

[0059] (4) Obtain the design drawings of the bridge to be inspected and preprocess them, input the preprocessed design drawings of the bridge to be inspected into the trained bridge structure detection model, and the trained bridge structure detection model outputs the image after the bridge multi-view structure target detection, and marks the relative position of the bridge structure target on the image after the bridge multi-view structure target detection; the relative position information of the identified target is given in the image, so that the target area can be accurately intercepted, avoiding the complexity of manual cropping and realizing an automated process.

[0060] If there is a bridge structure target that needs to be identified in the input image, the model will output a recognition result image with a target detection box and the relative position information of the detection box in the image; if it is determined that there is no bridge structure target that meets its recognition conditions in the input image, the model will output the original unlabeled image and output the prompt statement "no detections".

[0061] (5) Based on the image of the bridge after multi-view structural target detection, intercept the bridge structure image under different views, wherein the different views include a cross-sectional view, an elevation view, a top plan view, and a bottom plan view;

[0062] (6) Select different key parameters to be identified based on the bridge structure images under different views;

[0063] (7) Input the bridge structure images under different views and the different key parameters to be identified into the paddleocr model for feature extraction and feature fusion to obtain multimodal features. The multimodal features are input into the vi_layoutxlm_ser model for prediction to obtain an editable Excel document.

[0064] (8) Read the editable Excel document, and merge the editable Excel document of the key parameters of each view into a new Excel document of the key parameter information of the bridge structure in the same span according to the logical relationship between the parameters of multiple views of the bridge structure in the same span.

[0065] To identify different bridge structures from bridge design drawings, the YOLOv10 model was selected for multi-view recognition. YOLOv10 boasts fast detection speed and efficient real-time performance. Its advanced architecture, including a backbone network with strong feature extraction capabilities, a lightweight classification head, and spatial channel decoupling and downsampling, improves operational efficiency and generalization. Its lack of NMS training allows for object detection in complex scenarios, improving recognition accuracy and reliability. Its high-precision recognition capabilities, flexible model sizes to accommodate varying resource and accuracy requirements, and excellent scalability facilitate fine-tuning for specific tasks, thus facilitating accurate bridge structure recognition.

[0066] The step (1) specifically includes the following steps in order:

[0067] (1a) Use the pdf2image open source library to convert archived PDF bridge design drawings into JPG images. The resolution of each JPG image is 3573 × 2526.

[0068] (1b) Perform data augmentation by rotating all JPG images and adding Gaussian noise;

[0069] (1c) Labeling: Use the Labelimg tool to label the bridge structure in multiple views of the data augmented image, which are divided into cross-section, elevation, top plan, and bottom plan. The labeling format is YOLO format.

[0070] The step (2) specifically refers to: improving the YOLOv10 model, replacing all C2f modules in the YOLOv10 model with the DualConv module; and replacing the detection head in the YOLOv10 model with the detection head of the RT-DETR model.

[0071] The DualConv module is an innovative convolutional network architecture designed to build lightweight deep neural networks. It combines the advantages of 3*3 and 1*1 convolution kernels by processing the same input feature map channels. The 3*3 convolution kernel captures more spatial information during feature extraction, while the 1*1 convolution kernel facilitates interaction and information integration between feature channels without increasing parameters or computational complexity. The DualConv module also utilizes group convolution to efficiently arrange convolution filters. In group convolution, the input and output feature maps are divided into multiple groups, with the convolution filters in each group processing only a portion of the corresponding input feature map, reducing model complexity. Because different convolution kernels within a group (e.g., 3*3 and 1*1) can process the same set of input channels in parallel, information flow and feature extraction efficiency are optimized while maintaining the network's representational power and further reducing computational cost.

[0072] The RT-DETR model's detection head utilizes a more advanced and efficient feature fusion method, meticulously integrating features at different scales and levels, enabling the model to extract more comprehensive and accurate target features. Furthermore, compared to the YOLOv10 detection head, the RT-DETR model's detection head is more lightweight in design. This means it significantly reduces the model's parameter count and computational complexity while maintaining excellent detection performance, significantly improving detection efficiency. Furthermore, the RT-DETR model's detection head possesses exceptional multimodal information fusion capabilities. In addition to conventional visual features, it can also integrate information from other modalities, such as depth and semantics, to further clarify the target's category, greatly enhancing the model's ability to detect and recognize targets in complex scenarios.

[0073] The step (5) specifically includes the following steps in sequence:

[0074] (5a) Obtain the target frame information in the image after the bridge multi-view structure target detection, that is, the relative position of the bridge structure target in the image. The target frame information includes:<class_label><x_center><y_center> <width>and <height>;<class_label> Refers to the target category label, which uses an integer to represent the category number to which the detected target belongs;<x_center> and<y_center> Refers to the x-coordinate and y-coordinate of the center point of the target frame respectively; <width>and <height>Refers to the width and height of the target box respectively;

[0075] (5b) Convert the target frame information into coordinate information according to the following conversion relationship:

[0076] left = int((x_center - width / 2) * image.shape[1])

[0077] top = int((y_center - height / 2) * image.shape[0])

[0078] right = int((x_center + width / 2) * image.shape[1])

[0079] bottom = int((y_center + height / 2) * image.shape[0])

[0080] In the formula, image.shape[1] represents the width of the image, image.shape[0] represents the height of the image, left represents the horizontal coordinate position of the left boundary of the target box in the image; right represents the horizontal coordinate position of the right boundary of the target box in the image; top represents the vertical coordinate position of the upper boundary of the target box in the image; bottom represents the vertical coordinate position of the lower boundary of the target box in the image;

[0081] For each target, a Python script was used to capture the detected bridge structure image after multi-view target detection based on the target's coordinate information. The bridge structure images under different views, namely, cross-section, elevation, top plan, and bottom plan, were captured. The captured bridge structure images under different views were named "original image name + view name" and saved in the corresponding folders for each view.

[0082] The step (6) specifically refers to:

[0083] like Figure 2 As shown in the figure, the key parameters that need to be identified to determine the cross-section include: serial number, box beam upper flange width, box beam lower flange width, flange left cantilever width, flange right cantilever width, beam height, cantilever end thickness and cantilever root thickness, a total of 8 parameters;

[0084] like Figure 3 As shown, the key parameters that need to be identified in determining the elevation drawing include: section symbol, pier number, span and beam height, a total of 4 parameters;

[0085] like Figure 4 As shown, the key parameters that need to be identified to determine the top plan include: pier number, span and beam height, a total of 3 parameters;

[0086] like Figure 5 As shown, the key parameters that need to be identified to determine the bottom plan include: pier number, span and beam height, a total of 3 parameters.

[0087] As the blueprint guiding bridge construction, bridge design drawings encompass a vast array of parameters, encompassing every aspect of the bridge, including but not limited to structural component dimensions, material properties, mechanical performance indicators, and various structural details. However, in real-world workflows, specific tasks often require only a subset of key parameters. Indiscriminately identifying and extracting all parameters would inevitably lead to an exponential increase in workload, consuming significant human, material, and time resources while also potentially inefficient and wasting resources. Therefore, a more sensible and appropriate strategy is to selectively select parameters that are most relevant to the task at hand, tailored to specific needs. This approach not only accurately meets practical work requirements, but also maximizes resource allocation, improves work efficiency, and ensures the efficiency and accuracy of bridge design drawing parameter identification and extraction.

[0088] The step (7) specifically includes the following steps in order:

[0089] (7a) Annotating the bridge structure images under different views, marking the key parameters on the bridge structure images under different views, and obtaining the annotated bridge structure images under different views;

[0090] (7b) The annotated bridge structure images under different views are input into the paddleocr model for feature extraction, and the text information, text coordinate information and layout feature information are extracted; then, the extracted text information, text coordinate information and layout feature information are encoded and converted, and fused into multimodal features; then, these multimodal features are input into the vi_layoutxlm_ser model for forward reasoning calculation, and the category, position and specific text content of each semantic entity in the image are predicted, and the recognition result image and text file are output. The recognition result image is as follows: Figure 6 、 Figure 7 、 Figure 8 、 Figure 9 shown.

[0091] Given the uniqueness and complexity of bridge design drawings, which contain rich textual information, including parameter information and names, this information is extremely valuable for parameter reference. Therefore, it is necessary to explore effective character recognition methods. PaddleOCR is an open-source, ultra-lightweight OCR system model developed by Baidu PaddlePaddle. It can be widely used in many fields, including document recognition, bill recognition, and digital reading. It provides detection, classification, and recognition capabilities, supporting multi-language recognition, including mixed recognition of Chinese, English, numbers, and other character types.

[0092] (7c) Processing the text file: The text file includes the image path and each text recognition result entry, which includes the transcription, bbox, points, pred_id and pred fields and their corresponding values; transcription represents the text content recognized in the image; bbox represents the bounding box coordinate information of the area where the recognized text is located, which is given in the form of a list, representing the upper left corner horizontal coordinate, upper left corner vertical coordinate, lower right corner horizontal coordinate, and lower right corner vertical coordinate in order; points is used to describe the outline coordinate information of the area where the text is located, which is presented in the form of polygon vertex coordinates, and each element is a list containing horizontal and vertical coordinates; pred_id represents the number of the prediction result, and different numbers correspond to different recognition categories; pred represents the key parameter name corresponding to the recognized text, and the unlabeled text category is "NONE";

[0093] Use a Python script to filter all entries with a pred value other than "NONE" and the "transcription" entry within the dictionary containing that entry, and generate an editable Excel document. The summary document only requires the key parameter name, category, and content. Simply summarizing all output results will result in duplicate key parameters across subviews, causing data redundancy and increasing the workload for subsequent information processing and maintenance. Therefore, it is necessary to filter out irrelevant information.

[0094] The step (8) specifically includes the following steps in order:

[0095] (8a) The logical relationship between the parameters of multiple views of the same span bridge structure includes:

[0096] The elevation, top plan and bottom plan are identified by the pier number to which they belong. Piers with the same pier number belong to the same span.

[0097] The relationship between the serial number of the cross-section drawing and the section symbol number in the elevation drawing needs to be analyzed; within the same span, the serial number of the cross-section drawing must fall within the range of the section symbol number between the two pier numbers in the elevation drawing; for example, if the section symbol number range of the elevation drawing is 1 to 4, then the serial number of the cross-section drawing in the same span should be 1-1 to 4-4.

[0098] (8b) Based on the logical relationship between the parameters of multiple views of the bridge structure within the same span, the editable Excel documents are merged and summarized into a new Excel document containing the key parameter information of the bridge structure within the same span to realize parameter extraction.

[0099] Cast-in-place box girder design drawings are often complex, and identifying the design parameters for the entire bridge can be challenging, especially when the design involves extensive details and complex structures. In bridge design and engineering terminology, the entire bridge is typically divided into spans, with each span representing the distance between two piers (or supports). Therefore, during the parameter identification process, the entire bridge can be identified individually, breaking down the larger task into smaller pieces. The identification of each span is relatively independent, reducing complexity and potential for confusion, and ensuring data accuracy and integrity. Span-by-span identification also improves work efficiency, as identification of each span can be performed in parallel without having to wait for the entire bridge's data to be identified. Data management after span-by-span identification is simpler and more straightforward. Data for each span can be independently archived and managed, and updates can be processed individually without impacting the management and maintenance of the overall dataset.

[0100] like Figure 10 As shown, Figure 10 This diagram shows all the key parameters for the first span of a cast-in-place box girder. Section symbols 1, 2, 3, and 4 are present, along with pier numbers 0 and 1. This span has a length of 2495 cm and a beam height of 160 cm. Each section symbol corresponds to a cross-section, for a total of four cross-sections. The width of the upper flange and the lower flange of the box girder of section 1-1 are 850.1 cm and 450.1 cm respectively, the width of the upper flange and the lower flange of the box girder of section 2-2 are 853.2 cm and 453.2 cm respectively, the width of the upper flange and the lower flange of the box girder of section 3-3 are 874.9 cm and 474.8 cm respectively, the width of the upper flange and the lower flange of the box girder of section 4-4 are 903.1 cm and 503.1 cm respectively; the thickness of the cantilever end of section 1-1 is 50 cm, and the thickness of the cantilever ends of the other three sections is 20 cm; the thickness of the cantilever root, the cantilever width on the left side of the flange and the cantilever width on the right side of the flange of each section are the same, which are 50 cm, 200.1 cm and 200 cm respectively.

[0101] In summary, the present invention uses the improved YOLOv10 model to locate and intercept the image after the multi-view structure target detection of the bridge, and intercepts the bridge structure images under different views, namely the cross-section, elevation, top plan and bottom plan; uses the paddleocr model and the vi_layoutxlm_ser model to extract the position and value of the key parameters that need to be identified in each view; filters the content in the text file of the recognition result according to specific conditions, and finally integrates the Excel document of the same cross-key parameters according to the logical relationship of the parameters between multiple views; the present invention replaces the original structure corresponding to the YOLOv10 model with the DualConv module and the RT-DETR detection head, which can not only reduce parameters and improve the model's extraction accuracy of the features of components of different scales in the bridge design drawing, but also can combine Y The real-time detection capability of the YOLOv10 model and the deep feature understanding capability of the RT-DETR detection head; the present invention also assigns semantic information to the key parameters in the drawings, so that the parameter names and values ​​correspond to each other, realizes the modularization of the parameter identification of each plan view, and can efficiently and automatically extract the key parameters of the PDF drawings; finally, the method of identification and integration according to span is adopted to successfully solve the parameter identification problem of multiple views, and conducts useful exploration for the understanding of engineering design drawings by artificial intelligence algorithms; experimental results show that the accuracy and recall rate of the improved YOLOv10 model on the multi-view test set are 94.2% and 95.6% respectively; the results of the key parameters identified by the key information extraction method are basically consistent with the manual identification results; the present invention can quickly and accurately extract the key parameters of multiple views in the PDF drawings of cast-in-place box girders.< / height> < / width> < / height> < / width> < / height> < / width> < / height> < / width>

Claims

1. A method for identifying cross-key parameters of cast-in-situ box girder images based on target detection and key information extraction, characterized by: The method comprises the following steps in sequence: (1) Obtaining bridge design drawings and preprocessing them to obtain preprocessed bridge design drawings; (2) Improve the YOLOv10 model to obtain a bridge structure detection model; (3) Inputting the preprocessed bridge design drawings into the bridge structure detection model for training to obtain a trained bridge structure detection model; (4) Obtain the design drawings of the bridge to be inspected and preprocess them, input the preprocessed design drawings of the bridge to be inspected into the trained bridge structure detection model, and the trained bridge structure detection model outputs an image after the multi-view structure target detection of the bridge, and the relative position of the bridge structure target is marked on the image after the multi-view structure target detection of the bridge; (5) Based on the image of the bridge after multi-view structural target detection, intercept the bridge structure image under different views, wherein the different views include a cross-sectional view, an elevation view, a top plan view, and a bottom plan view; (6) Select different key parameters to be identified based on the bridge structure images under different views; (7) Input the bridge structure images under different views and the different key parameters to be identified into the paddleocr model for feature extraction and feature fusion to obtain multimodal features. The multimodal features are input into the vi_layoutxlm_ser model for prediction to obtain an editable Excel document. (8) Read the editable Excel document, and merge the editable Excel document of the key parameters of each view into a new Excel document of the key parameter information of the bridge structure in the same span according to the logical relationship between the parameters of multiple views of the bridge structure in the same span; The step (2) specifically refers to: improving the YOLOv10 model, replacing all C2f modules in the YOLOv10 model with the DualConv module; replacing the detection head in the YOLOv10 model with the detection head of the RT-DETR model; The step (7) specifically includes the following steps in order: (7a) Annotating the bridge structure images under different views, marking the key parameters on the bridge structure images under different views, and obtaining the annotated bridge structure images under different views; (7b) Input the annotated bridge structure images under different views into the PaddleOCR model for feature extraction, and extract text information, text coordinate information and layout feature information; The extracted text information, text coordinate information, and layout feature information are then encoded and transformed into multimodal features. These multimodal features are then fed into the vi_layoutxlm_ser model for forward reasoning, predicting the category, location, and specific text content of each semantic entity in the image. The recognition result image and text file are then output. (7c) Processing the text file: The text file includes the image path and each text recognition result entry, which includes the transcription, bbox, points, pred_id and pred fields and their corresponding values; transcription represents the text content recognized in the image; bbox represents the bounding box coordinate information of the area where the recognized text is located, given in the form of a list, representing the upper left corner horizontal coordinate, upper left corner vertical coordinate, lower right corner horizontal coordinate, and lower right corner vertical coordinate in order; points is used to describe the outline coordinate information of the area where the text is located, presented in the form of polygon vertex coordinates, and each element is a list containing horizontal and vertical coordinates; pred_id represents the number of the prediction result. Different numbers correspond to different recognition categories. pred indicates the key parameter name of the corresponding recognized text. The unlabeled text category is "NONE"; Use the Python script to filter out all entries whose pred value is not "NONE" and the "transcription" entry in the dictionary where the entry is located, and generate an editable Excel document.

2. The method for identifying cross-key parameters of cast-in-situ box girder images based on target detection and key information extraction according to claim 1 is characterized in that: The step (1) specifically includes the following steps in order: (1a) Use the pdf2image open source library to convert archived PDF bridge design drawings into JPG images. The resolution of each JPG image is 3573 × 2526. (1b) Perform data augmentation by rotating all JPG images and adding Gaussian noise; (1c) Labeling: Use the Labelimg tool to label the bridge structure in multiple views of the data augmented image, which are divided into cross-section, elevation, top plan, and bottom plan. The labeling format is YOLO format.

3. The method for identifying cross-key parameters of cast-in-situ box girder images based on target detection and key information extraction according to claim 1 is characterized in that: The step (5) specifically includes the following steps in sequence: (5a) Obtain the target frame information in the image after the bridge multi-view structure target detection, that is, the relative position of the bridge structure target in the image. The target frame information includes:<class_label><x_center><y_center> <width>and <height>;<class_label> Refers to the target category label, which uses an integer to represent the category number to which the detected target belongs;<x_center> and<y_center> Refers to the x-coordinate and y-coordinate of the center point of the target frame respectively; <width>and <height> Refers to the width and height of the target box respectively;< / height> < / width> < / height> < / width> (5b) Convert the target frame information into coordinate information according to the following conversion relationship: left = int((x_center - width / 2) * image.shape[1]) top = int((y_center - height / 2) * image.shape[0]) right = int((x_center + width / 2) * image.shape[1]) bottom = int((y_center + height / 2) * image.shape[0]) In the formula, image.shape[1] represents the width of the image, image.shape[0] represents the height of the image, left represents the horizontal coordinate position of the left boundary of the target box in the image; right represents the horizontal coordinate position of the right boundary of the target box in the image; top represents the vertical coordinate position of the upper boundary of the target box in the image; bottom represents the vertical coordinate position of the lower boundary of the target box in the image; For each target, a Python script was used to capture the detected bridge structure image after multi-view target detection based on the target's coordinate information. The bridge structure images under different views, namely, cross-section, elevation, top plan, and bottom plan, were captured. The captured bridge structure images under different views were named "original image name + view name" and saved in the corresponding folders for each view.

4. The method for identifying cross-key parameters of cast-in-situ box girder images based on target detection and key information extraction according to claim 1 is characterized in that: The step (6) specifically refers to: The key parameters that need to be identified to determine the cross-section include: serial number, box beam upper flange width, box beam lower flange width, flange left cantilever width, flange right cantilever width, beam height, cantilever end thickness and cantilever root thickness, a total of 8 parameters; The key parameters that need to be identified in the elevation drawing include: section symbol, pier number, span and beam height, a total of 4 parameters; The key parameters that need to be identified to determine the top plan include: pier number, span and beam height, a total of 3 parameters; The key parameters that need to be identified to determine the bottom plan include: pier number, span and beam height, a total of 3 parameters.

5. The method for identifying cross-key parameters of cast-in-situ box girder images based on target detection and key information extraction according to claim 1 is characterized in that: The step (8) specifically includes the following steps in order: (8a) The logical relationship between the parameters of multiple views of the same span bridge structure includes: The elevation, top plan and bottom plan are identified by the pier number to which they belong. Piers with the same pier number belong to the same span. The relationship between the serial number of the cross-section drawing and the cut symbol number in the elevation drawing needs to be analyzed; within the same span, the serial number of the cross-section drawing must fall within the cut symbol number range between two pier numbers in the elevation drawing; (8b) Based on the logical relationship between the parameters of multiple views of the bridge structure within the same span, the editable Excel documents are merged and summarized into a new Excel document containing the key parameter information of the bridge structure within the same span to realize parameter extraction.

Images