A communication engineering design CAD drawing auditing method and device

Abstract

This application provides a method and apparatus for reviewing CAD drawings used in communication engineering design. The method includes: parsing the CAD drawing to be reviewed to obtain title elements, graphic element elements, text elements, and table elements; generating title data, graphic element data, text data, and table data according to a preset data format; performing feature matching on the title data, graphic element data, text data, and table data to obtain a title information table, graphic element information table, text information table, and table information table; integrating the title information table, graphic element information table, text information table, and table information table in a structured manner based on the relationships between the elements of the CAD drawing to obtain structured data; generating an review rule engine based on the review rules; and reviewing the structured data using the review rule engine to obtain a review result. This method can improve review efficiency, reduce costs, and minimize human error.

Description

Technical Field

[0001] This application relates to the field of communication engineering technology, and in particular to a method and apparatus for reviewing CAD drawings for communication engineering design. Background Technology

[0002] Existing CAD drawings for telecommunications engineering designs are primarily in DWG and PDF formats. Currently, quality review of these drawings typically relies on manual review. Reviewers need extensive professional knowledge and experience to identify elements such as lines, nodes, and equipment in the drawings and check their compliance with design specifications and standards. This process is not only time-consuming but also susceptible to human error, leading to inconsistencies and errors in the review results. Furthermore, due to the sheer volume of drawings—for example, according to incomplete statistics, each province of China Mobile generates approximately 40,000 design drawings annually—a significant amount of manpower is required for review, resulting in substantial labor costs. Summary of the Invention

[0003] In view of this, this application provides a method and apparatus for reviewing CAD drawings for communication engineering design to solve the above-mentioned technical problems.

[0004] In a first aspect, embodiments of this application provide a method for reviewing CAD drawings used in communication engineering design, including: The CAD drawings to be reviewed are parsed to obtain title elements, graphic element elements, text elements, and table elements; Based on the image tag element, image element element, text element, and table element, generate image tag data, image element data, text data, and table data respectively according to the preset data format; Feature matching is performed on the image tag data, image element data, text data, and table data respectively to obtain an image tag information table containing multiple image tag data items, an image element information table containing multiple image element data items, a text information table containing multiple text data items, and a table information table containing multiple table data items. Based on the relationships between the elements of the CAD drawing, the title information table, element information table, text information table, and table information table are structurally integrated to obtain structured data; Based on the key points and standards of drawing review, review rules are formulated, and an review rule engine is generated based on the review rules. The structured data is reviewed using an audit rule engine to obtain the audit results. Secondly, embodiments of this application provide a device for reviewing CAD drawings used in communication engineering design, comprising: The parsing unit is used to parse the CAD drawings to be reviewed, and obtain the label elements, graphic element elements, text elements and table elements; The first generation unit is used to generate image tag data, image tag data, text data, and table data respectively based on image tag elements, image element elements, text elements, and table elements, according to a preset data format. The processing unit is used to perform feature matching on the image tag data, image element data, text data and table data respectively to obtain an image tag information table containing multiple image tag data items, an image element information table containing multiple image element data items, a text information table containing multiple text data items and a table information table containing multiple table data items. The integration unit is used to perform structured integration of the title information table, element information table, text information table and table information table based on the relationship between the elements of the CAD drawing to obtain structured data; The second generation unit is used to formulate review rules based on the key points and standards of drawing review, and generate a review rule engine based on the review rules; The audit unit is used to audit the structured data using an audit rule engine and obtain audit results.

[0005] Thirdly, embodiments of this application provide an electronic device, including: a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor executes the computer program to implement the method of embodiments of this application.

[0006] Fourthly, embodiments of this application provide a computer-readable storage medium storing computer instructions that, when executed by a processor, implement the methods of embodiments of this application.

[0007] Fifthly, embodiments of this application provide a computer program product, including a computer program / instructions, which, when executed by a processor, implement the method of embodiments of this application.

[0008] The method described in this application can improve review efficiency, reduce costs, minimize human error, enhance adaptability to complex and diverse communication engineering designs, handle various new design requirements and changes, and ensure the consistency and accuracy of the review. Attached Figure Description

[0009] To more clearly illustrate the technical solutions in the specific embodiments of this application or the prior art, the drawings used in the description of the specific embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are some embodiments of this application. For those skilled in the art, other drawings can be obtained from these drawings without creative effort.

[0010] Figure 1A flowchart illustrating the method for reviewing CAD drawings for communication engineering design provided in this application embodiment; Figure 2 Functional structure diagram of the communication engineering design CAD drawing review device provided in the embodiments of this application; Figure 3 This is a structural diagram of an electronic device provided in an embodiment of this application. Detailed Implementation

[0011] To make the objectives, technical solutions, and advantages of the embodiments of this application clearer, the technical solutions of the embodiments of this application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of this application, and not all embodiments. The components of the embodiments of this application described and shown in the accompanying drawings can generally be arranged and designed in various different configurations.

[0012] Therefore, the following detailed description of the embodiments of this application provided in the accompanying drawings is not intended to limit the scope of the claimed application, but merely to illustrate selected embodiments of the application. All other embodiments obtained by those skilled in the art based on the embodiments of this application without inventive effort are within the scope of protection of this application.

[0013] First, a brief introduction to the design concept of the embodiments of this application will be given.

[0014] Existing technologies have significant limitations in the scenario of reviewing communication engineering drawings: general drawing recognition technologies focus on the construction and industrial fields and have not been optimized for the unique complexity of line topology, diversity of equipment symbols and cross-disciplinary collaboration requirements of communication engineering, and cannot meet the recognition requirements of key elements in communication engineering drawings; moreover, traditional OCR technology can only recognize text characters and cannot parse the design logic in communication drawings.

[0015] Existing CAD drawings for communication engineering design are subject to manual review and have the following characteristics: Highly dependent on human intervention: The review process relies primarily on the experience and judgment of professionals.

[0016] Low efficiency: Manual review is slow and cannot quickly process a large number of drawings.

[0017] Error-prone: Human factors may lead to errors and inconsistencies in the audit results.

[0018] High cost: It requires a large investment of human resources, which increases the cost of review.

[0019] This application aims to address the shortcomings of the existing technology by proposing an intelligent review method for CAD drawings in communication engineering. The method includes: parsing the CAD drawing to be reviewed to obtain title elements, graphic element elements, text elements, and table elements; generating title data, graphic element data, text data, and table data according to preset data formats based on the title elements, graphic element elements, text elements, and table elements; performing feature matching on the title data, graphic element data, text data, and table data to obtain a title information table containing multiple title data items, a graphic element information table containing multiple graphic element data items, a text information table containing multiple text data items, and a table information table containing multiple table data items; structurally integrating the title information table, graphic element information table, text information table, and table information table based on the relationships between the elements of the CAD drawing to obtain structured data; formulating review rules based on the key points and standards of drawing review, and generating a review rule engine based on the review rules; and using the review rule engine to review the structured data to obtain the review result. The advantages of this application include: Full element coverage and domain adaptation: This proposal focuses on the review of CAD drawings (including DWG and PDF formats) in the field of communication engineering design, and solves the problems of insufficient recognition of communication drawing elements and poor rule adaptability of traditional general review technology, significantly improving the professionalism and domain adaptability of the review.

[0020] Full automation: Compared to manual or semi-automatic review processes, this proposal achieves full automation from data acquisition to report generation.

[0021] Advanced recognition algorithm: By introducing an analytical recognition model and utilizing multi-scale feature pyramids, graph attention mechanisms, and spatial-semantic fusion strategies, the recognition accuracy of elements such as lines, nodes, and equipment in communication engineering drawings is greatly improved compared to existing technologies, reducing misjudgments and omissions.

[0022] A systematic and sophisticated data processing workflow: Specific processing methods are adopted for different formats in data preprocessing, and comprehensive data cleaning, filtering and correlation are carried out in data analysis to make the data more accurate, effective and closely integrated with drawings. Existing technologies are not perfect and detailed enough in these aspects.

[0023] Flexible rule configuration: The rule engine supports zero-code configuration, allowing users to intuitively define and modify audit rules through a graphical interface to adapt to different audit needs and standards. Existing technologies are somewhat lacking in terms of flexibility and ease of use in rule configuration.

[0024] Detailed and automated audit report generation: This feature automatically generates detailed audit reports including an overall assessment of drawings, a list of issues, and improvement suggestions, providing users with comprehensive and valuable audit results. Existing technologies lag behind in terms of report comprehensiveness and automation.

[0025] After introducing the application scenarios and design concepts of the embodiments of this application, the technical solutions provided by the embodiments of this application will be described below.

[0026] like Figure 1 As shown, this application provides a method for reviewing CAD drawings used in communication engineering design, including the following steps: Step 101: Parse the CAD drawings to be reviewed to obtain label elements, graphic element elements, text elements, and table elements; Step 102: Based on the tag element, graphic element, text element, and table element, generate tag data, graphic element data, text data, and table data respectively according to the preset data format; Step 103: Perform feature matching on the image tag data, image element data, text data, and table data respectively to obtain an image tag information table containing multiple image tag data items, an image element information table containing multiple image element data items, a text information table containing multiple text data items, and a table information table containing multiple table data items. Step 104: Based on the relationships between the elements of the CAD drawing, the title information table, element information table, text information table, and table information table are structurally integrated to obtain structured data; Step 105: Develop review rules based on the key points and standards of drawing review, and generate a review rule engine based on the review rules; Step 106: The structured data is audited using the audit rule engine to obtain the audit results. The method in this embodiment can improve review efficiency, reduce costs, reduce human error, enhance adaptability to complex and diverse communication engineering designs, handle various new design requirements and changes, and ensure the consistency and accuracy of the review.

[0027] In some embodiments, the multiple attributes of the title element include: project name, drawing number, date, font type, font size, and position; the multiple attributes of the graphic element include: shape, line type, color, line style, line width, and position; the multiple attributes of the text element include: font, size, and rotation angle; and the multiple attributes of the table element include: equipment parameter list.

[0028] In some embodiments, the CAD drawing to be reviewed is converted into a DWG file; The CAD drawings to be reviewed are parsed to obtain title elements, graphic element elements, text elements, and table elements; including: Parse the structure of a DWG file, including header information, tables, and blocks; Extract the image tag element and its multiple attributes from the header information; The table is parsed to obtain the graphic elements and their multiple attributes; Extract graphic elements and their multiple attributes from the block; Extract the text content and annotations from the block to obtain the text element and its multiple attributes.

[0029] In some embodiments, the CAD drawing to be reviewed is converted into a PDF file; The CAD drawings to be reviewed are parsed to obtain title elements, graphic element elements, text elements, and table elements; including: Convert PDF files to images; The image is processed and enhanced to obtain the processed image; The pre-trained parsing and recognition model is used to process the image to obtain the tag elements and their multiple attributes, the primitive elements and their multiple attributes, the text elements and their multiple attributes, and the table elements and their multiple attributes.

[0030] Specifically, the image is processed and enhanced to obtain the processed image, including: Image denoising is performed to remove random noise and improve image clarity. By adjusting the image's grayscale histogram, dark areas are made darker and bright areas are made brighter, highlighting lines and symbols in the drawing; a sharpening algorithm is used to enhance the edges and details of the drawing frame, title block, and primitives.

[0031] In some embodiments, the parsing and recognition model includes: a multi-scale feature pyramid extraction module, a graph attention module, a spatial-semantic fusion module, and a fully connected layer; The method further includes: Establish a training set, which includes multiple CAD drawing image samples labeled with title elements, graphic elements, text elements, and table elements; Multi-scale feature maps of CAD drawing image samples are extracted using a multi-scale feature pyramid extraction module. The graph attention module is used to process the multi-scale feature map to obtain the spatial and semantic features of each node; The spatial and semantic features of each node are fused using the spatial-semantic fusion module to obtain the fused features of each node; The fusion features of each node are processed using a fully connected layer to obtain the node type and its multiple attributes; Calculate the loss value by combining the node type and its multiple attributes with the multiple attributes of the labeled elements; The parameters of the analytical recognition model are updated using the loss value.

[0032] Specifically, the collected drawing image samples are labeled, including the boundaries of the drawing frame (including line type, such as solid line, dashed line, line thickness, etc.), position (coordinate range on the page), and shape (circle, rectangle, irregular shape, etc.). For the title block, its content (including project name, drawing number, date, etc.), font type (SimSun, Heiti, etc.), font size, and position (coordinate range within the drawing frame) are labeled. Key elements such as lines (e.g., feeder lines, transmission lines), nodes (e.g., connection points, intersections), and equipment (e.g., base station equipment, transmission equipment) are labeled. The label content includes the element's category (e.g., specific base station model, transmission equipment type), location coordinates (accurate to the pixel level), attribute information (e.g., equipment power, line bandwidth), and the connection relationships between them.

[0033] A multi-scale feature pyramid is constructed using a series of convolutional kernels of different sizes and different pooling operations. For example, convolution operations are performed using 3x3, 5x5, and 7x7 kernels. The convolution operation can be represented as: For a 3x3 convolution kernel: F1 = Conv3x3(F) For a 5x5 convolution kernel: F2 = Conv5x5(F) For a 7x7 convolution kernel: F3 = Conv7x7(F) Where Conv represents the convolution operation and F represents the PDF image.

[0034] After convolution, feature maps F1, F2, and F3 at different scales are obtained. These feature maps contain information at different levels. The feature maps extracted by smaller convolution kernels focus more on local details, such as the subtle bends of the lines and the small component connections of nodes; the feature maps extracted by larger convolution kernels can better capture the global structure and patterns, such as the layout of the entire base station and the direction of the main lines.

[0035] Next, pooling operations are used to downsample the feature map to reduce its size while preserving the main features. Common pooling operations such as max pooling can be represented as: P1 = MaxPool(F1) P2 = MaxPool(F2) P3 = MaxPool(F3) Pooling is performed on the obtained feature map to reduce dimensionality and extract the main features.

[0036] By performing convolution and pooling operations at different scales, feature maps containing information at different levels are obtained, from local subtle features to overall macroscopic structure.

[0037] Each pixel or local region in the extracted multi-scale feature map is considered a node in the graph. The connections between nodes are determined based on spatial proximity, feature similarity, or prior topological knowledge.

[0038] Construct a graph G=(V,E), where V is the set of nodes and E is the set of edges.

[0039] Calculate the attention coefficients between nodes in the graph.

[0040] Attention coefficient between nodes i and j It can be calculated using the following formula:

[0041] in, and These are the feature vectors of nodes i and j. It is a linear transformation matrix. It is a learnable parameter vector, and LeakyReLU is the activation function. It is the set of neighboring nodes of node i.

[0042] Update the node's features based on the attention coefficient. New features of node i. It can be calculated as:

[0043] Where σ is the activation function.

[0044] By using graph attention mechanisms, the importance of key elements in the graph can be highlighted, and attention can be increased to important connections and structures.

[0045] Spatial features and semantic features are obtained separately. Spatial features can be geometric information about the shape, position, and size of the elements obtained through the preceding convolution and pooling operations; semantic features can be information about the element type and function obtained by encoding the labeled information.

[0046] Assuming the spatial features are represented as Its dimension is ds, and its semantic features are represented as Its dimension is d. Fusion is performed in the following manner:

[0047] in, These are learnable weight parameters that are trained to automatically adjust the relative importance of spatial and semantic features.

[0048] The fused features are input into a fully connected layer, where linear transformations and activation functions map the features to the final output class space. The output recognition result is a detailed description containing element categories, locations, attributes, and connectivity relationships. For example, for a base station device, the output might include device type, coordinate location, power parameters, and connectivity with other devices and lines.

[0049] This embodiment addresses the characteristics of communication engineering drawings (such as complex line topology and diverse equipment symbols) by designing a multi-scale feature pyramid. Using convolutional kernels of different sizes (3×3, 5×5, 7×7), it achieves hierarchical feature extraction of local details (such as line curvature angles) and global structure (such as base station layout). A graph attention mechanism is introduced, modeling drawing elements (lines, nodes, equipment) as graph structure nodes. Attention coefficients are used to calculate the association weights between nodes, enhancing the recognition accuracy of critical connections (such as the feeder and base station interface). A spatial-semantic fusion strategy is proposed, fusing geometric features (location, shape) and semantic features (equipment type, line function). For example, spatial features are used to locate the transmission line's direction, and semantic features are combined to determine whether it is a trunk optical cable.

[0050] In some embodiments, before feature matching is performed on the image caption data, image data, text data, and tabular data, each data item is filtered, including: The process of parsing and recognizing CAD drawings for communication engineering designs generates a wealth of graphic element information (such as graphic representations of lines, nodes, and equipment), text information (such as annotations and descriptions), tabular information (such as equipment parameter lists), and drawing signature information models (such as drawing name, number, and version). To facilitate subsequent intelligent verification, a series of filtering rules and standards are established to select effective information from this massive amount of data as the foundation. This can be based on the design specifications of the communication engineering project, project requirements, and common characteristics of effective information. For example, for text information, only descriptions related to key equipment parameters and technical indicators may be selected; for graphic element information, the focus may be on specific types of line connection methods and the graphic features of important nodes.

[0051] In some embodiments, feature matching is performed on the image tag data, image element data, text data, and table data to obtain an image tag information table containing multiple image tag data items, an image element information table containing multiple image element data items, a text information table containing multiple text data items, and a table information table containing multiple table data items; including: The features of each sign in the sign data are compared with the features of a predefined standard sign to obtain the sign data items; The features of each element in the image data are compared with the predefined standard element features to obtain the image data item; For example, for graphic data, its shape, size, position, and other features are analyzed and compared with predefined standard graphic element features. For instance, this can be used to determine whether the curvature of a line meets the required specifications.

[0052] Keywords or semantics are extracted from the features of text data, the keywords are matched with a pre-established keyword database, and the semantics are matched with predefined semantic patterns to obtain text data items. For example, check whether the text description accurately mentions the model and performance parameters of a specific device.

[0053] The table data items are obtained by comparing the characteristics of each table in the table data with the predefined standard table characteristics. For example, compare the table's structure, column names, data format, etc., with the expected standard table features; ensure that the data types, value ranges, etc. in the table meet the design requirements.

[0054] Associate the image data items, text data items, and table data items with the image label data to obtain the image label information table, image element information table, text information table, and table information table.

[0055] For example, the identified valid data is closely associated with drawing labels. By matching identifier fields or other related fields, such as drawing numbers and version numbers, a clear relationship is established between the data. This association allows for a clear understanding during the review process of the specific drawing to which each piece of valid data belongs, as well as its position and role in the overall design. Semantic relationships between elements, text, and tables are established through unique identifiers (such as associating equipment element IDs with model numbers in text annotations), solving the traditional OCR problem of "image-text separation."

[0056] The title block data items include: id, task ID, drawing ID, column number (left to right), row number (top to bottom), width, height, and number (format: F + row number (2 digits) + column number (2 digits)).

[0057] The table data items include: id, task ID, drawing ID, column number (from left to right), row number (from top to bottom), drawing frame number, table number, table name, row number, column number, cell text content, and rows and columns to be merged.

[0058] The graphic metadata items include: the column number of the drawing frame position, the row number of the frame position, the number of the drawing frame where it is located, the name of the feature type, the name of the block, the color, the name of the layer, the line width, the rotation angle, the name of the shape, the thickness, the width, the text size, the rotation angle of the block shape, the name of the drawing, the drawing frame size, the title block size, the drawing unit, the length and width of the graphic element, the area of the graphic element, the aspect ratio, the allowable deviation, and the coordinate of the graphic element position.

[0059] In some embodiments, based on the association relationships between the elements of the CAD drawing, the title block information table, the graphic element information table, the text information table, and the table information table are structurally integrated to obtain structured data, including: Obtain various information tables of the data source to be structurally integrated, and the data source includes: site construction, antenna feeder system, and bearing resources; Exemplarily, each data source contains multiple data fields. For example, the site construction data source may contain fields such as site construction type, site antenna hanging height, and site height.

[0060] Use the data structure integration rules to convert and process various information tables of the data source to be structurally integrated, and obtain an equipment information table and a performance information table. The equipment information table includes: a base station information table, an equipment list table, and a cable list table; the performance information table includes: a seismic intensity information table, a standard specification table, and a safety risk table.

[0061] Exemplarily, the data items of the cable list table in the structural integration result include: id, type, drawing attachment id, task id, title block, main workload table, main material table, and optical cable construction drawing.

[0062] Specifically, the review process includes: Establish a review rule engine; [[ END]]Exemplarily, for example: whether there is a routing diagram in the drawing: SELECT DISTINCT frame_no FROM prop_cad_table WHERE fileid = fileid AND TABLE_NAME = 'title block' AND cell_text LIKE '%routing diagram%' AND cell_text NOT LIKE '%and%', where fileid is the drawing file ID and frame_no is the drawing number. If the drawing number is not found, there is no routing diagram. If the drawing number is found, there is a routing diagram.

[0063] Audit rule execution: Based on the configured audit rules, the drawing data is actually audited and processed. During execution, every element and data item in the drawing is compared and judged according to the rules. For complex drawings, multi-level and multi-dimensional analysis and calculations are required to ensure the comprehensiveness and accuracy of the audit. For example, when auditing a communication network topology diagram with multiple levels, it is necessary not only to check whether the parameters of each node device are compliant, but also to determine whether the connection relationships between each level conform to the rules.

[0064] Audit rule verification: Once the drawing data has been processed according to the audit rules, it needs to be verified to ensure the accuracy and reliability of the audit.

[0065] Audit Result Recording: Record all results and operations during the audit process and store the audit results in the system database, including pass, fail, error messages, etc., to facilitate problem tracking and subsequent analysis.

[0066] Rule optimization and updates: Continuously optimize and update the review rules based on review results and user feedback.

[0067] User notifications and interactions: Promptly notify users of the review status and results during or after the review process.

[0068] In some embodiments, the method further includes: automatically generating a detailed audit report after completing the audit rule verification, including an overall assessment of the drawings, a list of issues, and improvement suggestions. Specifically, this includes: Summary information: Collects the audit results from the rules engine module, including all detected issues and elements that conform to the specifications.

[0069] Generate a draft report: Based on the summarized information, generate a preliminary audit report, including various formats such as text descriptions, tables, and charts, to clearly present the audit results.

[0070] Format the report: Format the draft report to ensure its professionalism and readability.

[0071] Publish the report: Publish the final audit report to the system for users to view.

[0072] Based on the same inventive concept, this application provides a device for reviewing CAD drawings used in communication engineering design. (See attached document.) Figure 2 As shown, the communication engineering design CAD drawing review device 200 provided in this application embodiment includes at least: Parsing unit 201 is used to parse the CAD drawings to be reviewed, and obtain the title elements, graphic elements, text elements and table elements; The first generation unit 202 is used to generate image tag data, image tag data, text data and table data respectively based on image tag elements, image element elements, text elements and table elements according to preset data formats; The processing unit 203 is used to perform feature matching on the image tag data, image element data, text data and table data respectively to obtain an image tag information table containing multiple image tag data items, an image element information table containing multiple image element data items, a text information table containing multiple text data items and a table information table containing multiple table data items. Integration unit 204 is used to perform structured integration of the title information table, element information table, text information table and table information table based on the relationship between the elements of the CAD drawing to obtain structured data; The second generation unit 205 is used to formulate review rules based on the key points and standards of drawing review, and generate a review rule engine based on the review rules; The audit unit 206 is used to audit the structured data using an audit rule engine to obtain audit results.

[0073] It should be noted that the principle of the communication engineering design CAD drawing review device 200 provided in this application embodiment to solve the technical problem is similar to the method provided in this application embodiment. Therefore, the implementation of the communication engineering design CAD drawing review device 200 provided in this application embodiment can refer to the implementation of the method provided in this application embodiment, and the repeated parts will not be described again.

[0074] Based on the same inventive concept, embodiments of this application also provide an electronic device, such as... Figure 3 As shown, it includes a memory and a processor. The memory stores an executable program, and the processor executes the executable program to implement the steps of the communication engineering design CAD drawing review method provided in the above embodiment.

[0075] The aforementioned processor can be a general-purpose processor, a digital signal processor, an application-specific integrated circuit (ASIC), a programmable logic device (PLD), or a combination thereof. The aforementioned PLD can be a complex programmable logic device (CPLD), a field-programmable gate array (FPGA), a generic array logic (GAL), or any combination thereof. The general-purpose processor can be a microprocessor or any conventional processor, etc.

[0076] Since the electronic device described in this application embodiment is an electronic device equipped with a memory for implementing the communication engineering design CAD drawing review method disclosed in this application embodiment, those skilled in the art can understand the structure and variations of the electronic device described in this application embodiment based on the communication engineering design CAD drawing review method described in this application embodiment, and therefore will not be described again here.

[0077] This application also provides a computer-readable storage medium storing a computer program thereon, which, when run by a processor, implements the steps of the communication engineering design CAD drawing review method provided in the above embodiments.

[0078] The storage medium in this embodiment may be included in an electronic device; or it may exist independently and not be assembled into an electronic device. The storage medium carries one or more computer programs, which, when executed, implement the steps of the communication engineering design CAD drawing review method provided in the above embodiment.

[0079] It should be understood that the various solutions in this embodiment have the same technical effects as those in the above method embodiments, and will not be repeated here.

[0080] According to embodiments of this application, the computer-readable storage medium can be a non-volatile computer-readable storage medium, such as including but not limited to: portable computer disks, hard disks, random access memory (RAM), read-only memory (ROM), erasable programmable read-only memory (EPROM or flash memory), portable compact disk read-only memory (CD-ROM), optical storage devices, magnetic storage devices, or any suitable combination thereof. Optionally, specific examples in this embodiment can refer to the examples described in any embodiment of this application, which will not be repeated here. Obviously, those skilled in the art should understand that the various modules or steps of this application described above can be implemented using general-purpose computing devices. They can be centralized on a single computing device or distributed across a network of multiple computing devices. Optionally, they can be implemented using computer-executable program code, thereby storing them in a storage device for execution by a computing device. In some cases, the steps shown or described can be performed in a different order than those presented here, or they can be fabricated as separate integrated circuit modules, or multiple modules or steps can be fabricated as a single integrated circuit module. Thus, this application is not limited to any particular hardware and software combination.

[0081] This application also provides a computer program product, including a computer program / instruction, which, when executed by a processor, implements the steps of the communication engineering design CAD drawing review method provided in the above embodiments.

[0082] The flowcharts and block diagrams in the accompanying drawings illustrate the architecture, functionality, and operation of possible implementations of systems, methods, and computer program products according to various embodiments of this application. In this regard, each block in a flowchart or block diagram may represent a module, segment, or portion of code containing one or more executable instructions for implementing a specified logical function. It should also be noted that in some alternative implementations, the functions targeted in the blocks may occur in a different order than those targeted in the drawings. For example, two consecutively represented blocks may actually be executed substantially in parallel, and they may sometimes be executed in reverse order, depending on the functions involved. It should also be noted that each block in the block diagrams and / or flowcharts, and combinations of blocks in the block diagrams and / or flowcharts, can be implemented using a dedicated hardware-based system that performs the specified function or operation, or using a combination of dedicated hardware and computer instructions.

[0083] Furthermore, while the operations are described in a specific order, this should not be construed as requiring these operations to be performed in the specific order shown or in a sequential order. Multitasking and parallel processing may be advantageous in certain environments. Similarly, while several specific implementation details are included in the above discussion, these should not be construed as limiting the scope of this application. Certain features described in the context of individual embodiments may also be implemented in combination in a single embodiment. Conversely, various features described in the context of a single embodiment may also be implemented individually or in any suitable sub-combination in multiple embodiments.

Claims

1. A method for reviewing CAD drawings used in communication engineering design, characterized in that, include: The CAD drawings to be reviewed are parsed to obtain title elements, graphic element elements, text elements, and table elements; Based on the image tag element, image element element, text element, and table element, generate image tag data, image element data, text data, and table data respectively according to the preset data format; Feature matching is performed on the image tag data, image element data, text data, and table data respectively to obtain an image tag information table containing multiple image tag data items, an image element information table containing multiple image element data items, a text information table containing multiple text data items, and a table information table containing multiple table data items. Based on the relationships between the elements of the CAD drawing, the title information table, element information table, text information table, and table information table are structurally integrated to obtain structured data; Based on the key points and standards of drawing review, review rules are formulated, and an review rule engine is generated based on the review rules. The structured data is reviewed using an audit rule engine to obtain the audit results.

2. The method according to claim 1, characterized in that, The title element has multiple attributes including: project name, drawing number, date, font type, font size, and position; the graphic element has multiple attributes including: shape, line type, color, line style, line width, and position; the text element has multiple attributes including: font, size, and rotation angle; and the table element has multiple attributes including: equipment parameter list.

3. The method according to claim 2, characterized in that, When the CAD drawing to be reviewed is converted into a DWG file; The CAD drawings to be reviewed are parsed to obtain title elements, graphic element elements, text elements, and table elements; including: Parse the structure of a DWG file, including header information, tables, and blocks; Extract the image tag element and its multiple attributes from the header information; The table is parsed to obtain the graphic elements and their multiple attributes; Extract graphic elements and their multiple attributes from the block; Extract the text content and annotations from the block to obtain the text element and its multiple attributes.

4. The method according to claim 2, characterized in that, When the CAD drawings to be reviewed are converted into PDF files; The CAD drawings to be reviewed are parsed to obtain title elements, graphic element elements, text elements, and table elements; including: Convert PDF files to images; The image is processed and enhanced to obtain the processed image; The pre-trained parsing and recognition model is used to process the image to obtain the tag elements and their multiple attributes, the primitive elements and their multiple attributes, the text elements and their multiple attributes, and the table elements and their multiple attributes.

5. The method according to claim 4, characterized in that, The analytical recognition model includes: a multi-scale feature pyramid extraction module, a graph attention module, a spatial-semantic fusion module, and a fully connected layer; The method further includes: Establish a training set, which includes multiple CAD drawing image samples labeled with title elements, graphic elements, text elements, and table elements; Multi-scale feature maps of CAD drawing image samples are extracted using a multi-scale feature pyramid extraction module. The graph attention module is used to process the multi-scale feature map to obtain the spatial and semantic features of each node; The spatial and semantic features of each node are fused using the spatial-semantic fusion module to obtain the fused features of each node; The fusion features of each node are processed using a fully connected layer to obtain the node type and its multiple attributes; Calculate the loss value by combining the node type and its multiple attributes with the multiple attributes of the labeled elements; The parameters of the analytical recognition model are updated using the loss value.

6. The method according to claim 1, characterized in that, Feature matching is performed on the image tag data, image element data, text data, and table data to obtain an image tag information table containing multiple image tag data items, an image element information table containing multiple image element data items, a text information table containing multiple text data items, and a table information table containing multiple table data items; including: The features of each sign in the sign data are compared with the features of a predefined standard sign to obtain the sign data items; The features of each element in the image data are compared with the predefined standard element features to obtain the image data item; Keywords or semantics are extracted from the features of text data, the keywords are matched with a pre-established keyword database, and the semantics are matched with predefined semantic patterns to obtain text data items. The table data items are obtained by comparing the characteristics of each table in the table data with the predefined standard table characteristics. Associate the image data items, text data items, and table data items with the image label data to obtain the image label information table, image element information table, text information table, and table information table.

7. The method according to claim 1, characterized in that, Based on the relationships between various elements in CAD drawings, the title information table, element information table, text information table, and tabular information table are structurally integrated to obtain structured data, including: Obtain various information tables of the data sources to be structured and integrated, including: site construction, antenna feeder system, and bearer resources; The data structure integration rules are used to transform and process various information tables of the data source to be structured and integrated, resulting in equipment information tables and performance information tables. The equipment information tables include: base station information tables, equipment list tables, and cable list tables; the performance information tables include: seismic intensity information tables, standard and specification tables, and safety risk tables.

8. The method according to claim 1, characterized in that, The audit results include: passed, failed, and error message.

9. A device for reviewing CAD drawings used in communication engineering design, characterized in that, include: The parsing unit is used to parse the CAD drawings to be reviewed, and obtain the label elements, graphic element elements, text elements and table elements; The first generation unit is used to generate image tag data, image tag data, text data, and table data respectively based on image tag elements, image element elements, text elements, and table elements, according to a preset data format. The processing unit is used to perform feature matching on the image tag data, image element data, text data and table data respectively to obtain an image tag information table containing multiple image tag data items, an image element information table containing multiple image element data items, a text information table containing multiple text data items and a table information table containing multiple table data items. The integration unit is used to perform structured integration of the title information table, element information table, text information table and table information table based on the relationship between the elements of the CAD drawing to obtain structured data; The second generation unit is used to formulate review rules based on the key points and standards of drawing review, and generate a review rule engine based on the review rules; The audit unit is used to audit the structured data using an audit rule engine and obtain audit results.

10. An electronic device, characterized in that, include: A memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor, when executing the computer program, implements the method as claimed in any one of claims 1-8.

Images