Construction drawing review optimization method and system based on deep learning

Abstract

The invention provides a construction drawing review optimization method and system based on deep learning, and the method comprises the steps: 1, obtaining a plurality of construction drawing samples, and carrying out the data annotation of the construction drawing samples, so as to obtain annotation samples; 2, constructing a deep learning model, and inputting the labeled sample into the deep learning model for training so as to construct a deep learning optimization model; 3, obtaining a to-be-processed construction drawing, preprocessing the to-be-processed construction drawing to obtain preprocessed data, and identifying and analyzing the preprocessed data by using the deep learning optimization model to extract key information in the preprocessed data; and 4, performing matching verification on the key information and a construction drawing review specification, and generating a corresponding optimization strategy based on a matching verification result.

Description

Technical Field

[0001] The present application relates to the field of data processing technology, and in particular to a construction drawing review optimization method and system based on deep learning. Background Art

[0002] Currently, the review of construction drawings is usually conducted offline. The responsible person prepares the project review plan and notifies the corresponding design unit to submit the materials. The design unit uploads the materials through the digital management center within its Economic Research Institute. After the submission, the Economic Research Institute organizes the corresponding experts to pre-review the materials. After the pre-review is completed, the time for the formal review meeting is arranged. During the formal review meeting, the professional review experts need to prepare meeting minutes, score the design quality score sheet, and finally complete the preparation of the review opinions.

[0003] The process is cumbersome and manual, requiring human involvement at every stage, from document upload and pre-review to meeting scheduling and drafting of review opinions. This is not only time-consuming and labor-intensive, but also susceptible to human factors, making it difficult to ensure review efficiency and accuracy. Furthermore, offline review methods are not conducive to information sharing and collaboration, limiting the optimal allocation of review resources. Summary of the Invention

[0004] The embodiments of the present application provide a construction drawing review optimization method and system based on deep learning to at least address the deficiencies in the above-mentioned related technologies.

[0005] In a first aspect, an embodiment of the present application provides a construction drawing review optimization method based on deep learning, comprising the following steps: Step 1: Obtain several construction drawing samples and annotate the construction drawing samples to obtain annotated samples; Step 2: Construct a deep learning model, and input the labeled samples into the deep learning model for training to construct a deep learning optimization model; Step 3: Obtain the construction drawing to be processed, and preprocess the construction drawing to be processed to obtain preprocessed data, and use the deep learning optimization model to identify and analyze the preprocessed data to extract key information from the preprocessed data; Step 4: Match and verify the key information with the construction drawing review specifications, and generate corresponding optimization strategies based on the matching verification results.

[0006] Furthermore, the step 1 includes: Collect construction drawing samples from various construction projects and obtain element types and corresponding positioning information of each construction drawing sample; Data annotation is performed on each of the construction drawing samples according to the element type and the positioning information to obtain a corresponding annotated sample.

[0007] Furthermore, the step three includes: Obtaining the construction drawings to be processed uploaded by the user, and converting the format of the construction drawings to be processed to obtain preliminary construction drawings; The preliminary construction drawing is resized to obtain corresponding preprocessed data, and the image classification algorithm and target detection algorithm of the deep learning optimization model are used to perform image classification and analysis detection on the preprocessed data to extract corresponding key information.

[0008] Furthermore, the step 4 includes: Match and verify the key information with the construction drawing review specifications, and mark data that does not comply with the construction drawing review specifications; A corresponding optimization strategy is generated from a preset strategy library based on the matching verification result, and the optimization strategy is fed back to the reviewer.

[0009] Furthermore, the method further comprises: Feedback opinions are collected at preset intervals to form a corresponding feedback opinion database; The deep learning optimization model is optimized and improved according to the feedback library to obtain a deep learning improved model.

[0010] In a second aspect, the present invention further proposes a construction drawing review and optimization system based on deep learning, comprising: A data annotation module is used to obtain a number of construction drawing samples and perform data annotation on the construction drawing samples to obtain annotated samples; A model training module is used to build a deep learning model and input the labeled samples into the deep learning model for training to build a deep learning optimization model; A data processing module is used to obtain the construction drawings to be processed, pre-process the construction drawings to be processed to obtain pre-processed data, and use the deep learning optimization model to identify and analyze the pre-processed data to extract key information from the pre-processed data; The strategy generation module is used to match and verify the key information with the construction drawing review specifications, and generate corresponding optimization strategies based on the matching verification results.

[0011] Furthermore, the data annotation module includes: A data acquisition unit is used to collect construction drawing samples from various construction projects and obtain the element type and corresponding positioning information of each construction drawing sample; The data annotation unit is used to perform data annotation on each of the construction drawing samples according to the element type and the positioning information to obtain a corresponding annotation sample.

[0012] Furthermore, the data processing module includes: A format conversion unit, configured to obtain a construction drawing to be processed uploaded by a user, and convert the format of the construction drawing to be processed to obtain a preliminary construction drawing; A data processing unit is used to resize the preliminary construction drawing to obtain corresponding preprocessed data, and use the image classification algorithm and target detection algorithm of the deep learning optimization model to perform image classification and analysis and detection on the preprocessed data to extract corresponding key information.

[0013] Furthermore, the strategy generation module includes: A matching and verification unit, configured to match and verify the key information with the construction drawing review specification, and mark data that does not comply with the construction drawing review specification; The strategy generation unit is used to generate a corresponding optimization strategy from a preset strategy library according to the matching verification result, and feed back the optimization strategy to the reviewer.

[0014] Furthermore, the system further comprises: The feedback collection module is used to collect feedback at preset intervals to form a corresponding feedback database; The optimization and improvement module is used to optimize and improve the deep learning optimization model according to the feedback library to obtain a deep learning improved model.

[0015] Compared with related technologies, the embodiment of the present application provides a construction drawing review optimization method and system based on deep learning, which uses deep learning technology to realize automatic recognition and analysis of drawings, greatly improves review efficiency, reduces the time and labor costs required for manual review, and can learn and accumulate review experience through deep learning models, continuously optimize review standards and processes, make review results more accurate and reliable, avoid omissions and errors caused by human factors in offline reviews, use online methods to solve time and space limitations, support remote synchronous office work, make review work more flexible and convenient, effectively solve the problems of low efficiency, poor accuracy and time and space limitations in offline reviews, and promote the digital transformation and intelligent development of construction drawing review work.

[0016] The details of one or more embodiments of the present application are set forth in the following drawings and description to make other features, objects, and advantages of the present application more readily apparent. BRIEF DESCRIPTION OF THE DRAWINGS

[0017] The drawings described herein are used to provide a further understanding of the present application and constitute a part of the present application. The illustrative embodiments of the present application and their descriptions are used to explain the present application and do not constitute an improper limitation on the present application. In the drawings: Figure 1 Flowchart of a construction drawing review optimization method based on deep learning in a first embodiment of the present invention; Figure 2 This is a structural block diagram of the construction drawing review and optimization system based on deep learning in the second embodiment of the present invention.

[0018] The following specific embodiments will further illustrate the present invention in conjunction with the above-mentioned drawings. DETAILED DESCRIPTION

[0019] In order to make the purpose, technical solutions and advantages of this application more clearly understood, the present application is described and illustrated below in conjunction with the accompanying drawings and examples. It should be understood that the specific embodiments described herein are merely used to explain this application and are not intended to limit this application. Based on the embodiments provided in this application, all other embodiments obtained by those of ordinary skill in the art without making any creative efforts are within the scope of protection of this application.

[0020] Obviously, the drawings described below are merely examples or embodiments of the present application. Those skilled in the art can, without inventive effort, apply the present application to other similar scenarios based on these drawings. Furthermore, it is also understood that, although the effort involved in such a development process may be complex and lengthy, for those skilled in the art related to the content disclosed in this application, changes in design, manufacturing, or production based on the technical content disclosed in this application are merely conventional technical means and should not be construed as an insufficiency of the content disclosed in this application.

[0021] References to "embodiments" in this application mean that a particular feature, structure, or characteristic described in connection with the embodiment may be included in at least one embodiment of the application. The appearance of this phrase in various places in the specification does not necessarily refer to the same embodiment, nor does it refer to independent or alternative embodiments that are mutually exclusive of other embodiments. It is understood, both explicitly and implicitly, by those skilled in the art that the embodiments described in this application may be combined with other embodiments unless there is a conflict.

[0022] Unless otherwise defined, the technical or scientific terms involved in this application should have the usual meaning understood by people with ordinary skills in the technical field to which this application belongs. The words "one", "a", "the" and the like involved in this application do not indicate a quantitative limitation and may indicate the singular or plural. The terms "include", "comprise", "have" and any variations thereof involved in this application are intended to cover non-exclusive inclusions; for example, a process, method, system, product or device that includes a series of steps or modules (units) is not limited to the listed steps or units, but may also include steps or units that are not listed, or may also include other steps or units that are inherent to these processes, methods, products or devices. The words "connect", "connected", "coupled" and the like involved in this application are not limited to physical or mechanical connections, but may include electrical connections, whether direct or indirect. Example 1

[0023] See also Figure 1 , which shows a construction drawing review optimization method based on deep learning in the first embodiment of the present invention, and the method specifically includes steps S101 to S104: S101, obtaining a number of construction drawing samples, and annotating the construction drawing samples to obtain annotated samples; Furthermore, the step S101 specifically includes steps S1011 to S1012: S1011, collecting construction drawing samples from various construction projects, and obtaining element types and corresponding positioning information of each construction drawing sample; S1012: Annotate each of the construction drawing samples according to the element type and the positioning information to obtain a corresponding annotated sample.

[0024] During implementation, construction drawing samples are collected from a variety of construction projects, covering different project types, to ensure the representativeness and comprehensiveness of the data. The original drawings need to be uniformly converted into a standardized format that can be processed by the model, including the classification and positioning information of elements such as structures, equipment, and pipelines. The annotation process includes two dimensions: classification and positioning: classification annotation assigns category labels to drawing elements, and positioning annotation marks the position of elements through bounding boxes or semantic segmentation. To improve efficiency, target detection and image classification algorithms are used to assist in annotation. For example, annotation results are initially generated based on a pre-trained model, and errors are corrected through manual review, forming a semi-automated process of "model prediction-manual correction-model iterative optimization". At the same time, annotation must strictly follow industry review specifications and ensure annotation quality through multi-level manual review and algorithm verification. The final annotation data supports the training and optimization of deep learning models, providing a basis for intelligent review and suggestion generation.

[0025] S102, constructing a deep learning model, and inputting the labeled samples into the deep learning model for training to construct a deep learning optimization model; During implementation, the core model architecture is built using the PyTorch or TensorFlow framework, serving as the "intelligent hub." Algorithms are used to precisely locate equipment, pipelines, and other elements in the drawings. The application of these models and algorithms enables effective identification and analysis of various elements in the drawings. This then leads to data training. During this process, a large amount of precisely annotated construction drawing data is collected, covering detailed information about the structures, equipment, pipelines, and other elements in the drawings. This data is fed into the model for repeated training. During training, the model continuously compares the predicted results with the actual annotations. Using an optimizer, the model parameters are dynamically adjusted to optimize the learning direction, allowing the model to gradually learn and memorize the features in the drawings until it can accurately identify all details. After training is complete, the specification verification phase begins. Construction specifications are digitally encoded and integrated into the system, acting like a "rulebook."

[0026] According to these coding standards, the input drawings are automatically checked for compliance. If any non-compliance is found in the drawings, such as pipeline spacing not meeting the required 0.5-meter standard, it will be immediately marked to prompt the reviewer to identify the problem in time.

[0027] In the actual application stage. Users upload construction drawings to be reviewed. After receiving the drawings, they are automatically analyzed and processed, and a detailed optimization report is generated based on the recognition results of the deep learning model and engineering practice experience. For example, in the case of unreasonable pipeline directions, specific suggestions for adjusting the direction are made. In addition, user feedback data is collected, and the model is continuously optimized and upgraded to continuously improve the system's review accuracy and the rationality of optimization suggestions, thereby achieving more efficient and accurate construction drawing review and optimization. The entire process, from drawing upload, intelligent analysis to result feedback, is completed online. It not only greatly shortens the time required for manual review and effectively reduces omissions caused by human factors, but also adapts to the ever-changing construction drawing review needs through continuous self-optimization. In specific implementation, a deep learning framework (such as TensorFlow or PyTorch) is used to build a model, and the labeled construction drawing samples are input into the model for training, so that the model can accurately identify each element and detail in the drawing.

[0028] S103: Acquire a construction drawing to be processed, preprocess the construction drawing to be processed to obtain preprocessed data, and use the deep learning optimization model to identify and analyze the preprocessed data to extract key information from the preprocessed data; Furthermore, the step S103 specifically includes steps S1031 and S1032: S1031, obtaining the construction drawing to be processed uploaded by the user, and converting the format of the construction drawing to be processed to obtain a preliminary construction drawing; S1032, resizing the preliminary construction drawing to obtain corresponding preprocessed data, and using the image classification algorithm and target detection algorithm of the deep learning optimization model to perform image classification and analysis detection on the preprocessed data to extract corresponding key information.

[0029] During the specific implementation, the user uploads the construction drawings to be reviewed to the system, and the system performs preprocessing operations such as format conversion and size adjustment. It uses the trained deep learning model to automatically identify and analyze the preprocessed drawings to extract key information from the drawings.

[0030] In this embodiment, after construction drawings are uploaded, an automated preprocessing process begins. A format conversion module standardizes original drawings in various formats (such as DWG, PDF, or scanned images) into high-resolution bitmap or vector formats. Dimension normalization and image enhancement are then performed to ensure the standardization and clarity of the input data. Subsequently, a multi-task model based on deep learning analyzes the preprocessed drawings, identifying key elements such as structural components, equipment locations, and pipeline layouts. Compliance is then verified item by item against a codified national / industry standard rule base (e.g., the "Regulations on the Preparation of Construction Engineering Design Documents"). When violations are detected, the problem areas are automatically annotated (e.g., by overlaying bounding boxes and masks), and a structured report (in JSON format) is generated, including specific parameter deviations (e.g., fire escape width is less than 10% of the standard value) and optimization suggestions (e.g., adjusting pipeline coordinates to meet compliance requirements). Reviewers can view the annotated drawings and reports in real time through a human-computer interface. Data is encrypted and stored throughout the process, supporting collaborative review by multiple individuals. This process increases the efficiency of traditional manual review by 3-5 times, with an accuracy rate of over 97% in identifying key items, significantly reducing the risk of human omissions and achieving standardization, intelligence, and process-based review of construction drawings.

[0031] S104: Match and verify the key information with the construction drawing review specification, and generate a corresponding optimization strategy based on the matching and verification result.

[0032] Furthermore, the step S104 specifically includes steps S1041 and S1042: S1041, matching and verifying the key information with the construction drawing review specification, and marking data that does not comply with the construction drawing review specification; S1042: Generate a corresponding optimization strategy from a preset strategy library based on the matching verification result, and feed back the optimization strategy to the reviewer.

[0033] During specific implementation, the recognition results will be matched and verified with the construction drawing review specifications, and the parts that do not meet the specifications will be marked and prompted. Based on the recognition results of the deep learning model and engineering experience, automatic optimization suggestions will be provided to reviewers.

[0034] In this embodiment, image classification, target detection and other technologies in deep learning are applied to classify and locate the structures, equipment, pipelines, etc. in the drawings, and comprehensively review the drawings; the deep learning model is combined with the construction drawing review specifications to automatically match and verify with the specifications, and mark and prompt the parts that do not meet the specifications.

[0035] Specifically, within the deep learning-based intelligent construction drawing review and optimization model, the alignment and verification of key information with construction drawing review specifications, as well as the implementation of optimization strategies, are crucial for ensuring construction drawing quality. At the outset, the deep learning model uses image classification and object detection techniques to accurately extract key information, including structure, equipment, and pipelines, from pre-processed construction drawings. This information is then meticulously compared with the corresponding construction drawing review specifications. Any discrepancies are immediately flagged and the reviewer is clearly informed of the violated specification clauses. Based on the alignment and verification results, specific optimization strategies are generated for each component. For structural defects, component dimensions that do not meet requirements, or an unreasonable reinforcement ratio, recommendations are provided for adjusting the structural layout, resizing components, or optimizing reinforcement methods. For equipment, if equipment selection is inappropriate, layout is inappropriate, or parameters do not meet standards, recommendations are provided for appropriate equipment models, adjustments to equipment location, or the addition of auxiliary equipment. For pipelines, if pipeline routing, spacing, or crossings do not meet specifications, recommendations are provided for rerouting, adjusting spacing, or adding protective measures. Reviewers can intuitively view these marked issues, generated optimization suggestions, and corresponding regulatory basis through a friendly human-computer interaction interface. Reviewers can evaluate and adjust the optimization suggestions based on their own professional knowledge and the actual situation of the project. If they agree with the suggestions, they can apply them directly; if they have different ideas, they can enter feedback, record and store these feedback. Finally, feedback from reviewers and other users is collected regularly, and the deep learning model is continuously optimized and upgraded in combination with technological developments and updates to industry standards. By continuously improving model algorithms and enriching training data, the accuracy of model recognition and the effectiveness of optimization strategies can be further improved, which can better adapt to complex and changing construction drawing review scenarios and provide more reliable and efficient support for construction drawing review work.

[0036] In some optional embodiments, the method further comprises: Feedback opinions are collected at preset intervals to form a corresponding feedback opinion database; The deep learning optimization model is optimized and improved according to the feedback library to obtain a deep learning improved model.

[0037] During the specific implementation, reviewers view the review results and optimization suggestions through the human-computer interaction interface, and can provide feedback and adjustments. Feedback from reviewers and other users is collected regularly to understand the system usage and existing problems. Based on user feedback and technological development, the deep learning model is continuously optimized and upgraded to improve the accuracy and efficiency of the system.

[0038] In summary, the construction drawing review optimization method based on deep learning in the above-mentioned embodiment of the present invention uses deep learning technology to realize automatic recognition and analysis of drawings, greatly improves the review efficiency, reduces the time and labor costs required for manual review, and can learn and accumulate review experience through the deep learning model, continuously optimize the review standards and processes, make the review results more accurate and reliable, avoid omissions and errors caused by human factors in offline reviews, use online methods to solve the limitations of time and space, support remote synchronous office, make the review work more flexible and convenient, effectively solve the problems of low efficiency, poor accuracy and time and space limitations in offline reviews, and promote the digital transformation and intelligent development of construction drawing review work. Example 2

[0039] On the other hand, the present invention also proposes a construction drawing review optimization system based on deep learning, please refer to Figure 2 , shown is a construction drawing review and optimization system based on deep learning in a second embodiment of the present invention, comprising: The data annotation module 11 is used to obtain a number of construction drawing samples and annotate the construction drawing samples to obtain annotated samples; Furthermore, the data annotation module 11 includes: A data acquisition unit is used to collect construction drawing samples from various construction projects and obtain the element type and corresponding positioning information of each construction drawing sample; The data annotation unit is used to perform data annotation on each of the construction drawing samples according to the element type and the positioning information to obtain a corresponding annotation sample.

[0040] A model training module 12 is used to construct a deep learning model and input the labeled samples into the deep learning model for training to construct a deep learning optimization model; The data processing module 13 is used to obtain the construction drawings to be processed, pre-process the construction drawings to be processed to obtain pre-processed data, and use the deep learning optimization model to identify and analyze the pre-processed data to extract key information from the pre-processed data; Furthermore, the data processing module 13 includes: A format conversion unit, configured to obtain a construction drawing to be processed uploaded by a user, and convert the format of the construction drawing to be processed to obtain a preliminary construction drawing; A data processing unit is used to resize the preliminary construction drawing to obtain corresponding preprocessed data, and use the image classification algorithm and target detection algorithm of the deep learning optimization model to perform image classification and analysis and detection on the preprocessed data to extract corresponding key information.

[0041] The strategy generation module 14 is used to match and verify the key information with the construction drawing review specifications, and generate a corresponding optimization strategy based on the matching and verification results.

[0042] Furthermore, the strategy generation module 14 includes: A matching and verification unit, configured to match and verify the key information with the construction drawing review specification, and mark data that does not comply with the construction drawing review specification; The strategy generation unit is used to generate a corresponding optimization strategy from a preset strategy library according to the matching verification result, and feed back the optimization strategy to the reviewer.

[0043] Furthermore, the system further comprises: The feedback collection module is used to collect feedback at preset intervals to form a corresponding feedback database; The optimization and improvement module is used to optimize and improve the deep learning optimization model according to the feedback library to obtain a deep learning improved model.

[0044] The functions or operation steps implemented when the above modules are executed are substantially the same as those in the above method embodiments and will not be described in detail here.

[0045] An embodiment of the present invention provides a construction drawing review and optimization system based on deep learning, the implementation principle and technical effects of which are the same as those of the aforementioned method embodiment. For the sake of brief description, for matters not mentioned in the system embodiment, please refer to the corresponding content in the aforementioned method embodiment.

[0046] The technical features of the above-mentioned embodiments can be combined arbitrarily. In order to make the description concise, not all possible combinations of the technical features in the above-mentioned embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this specification.

[0047] The above-described embodiments merely represent several implementation methods of the present application. While the descriptions are relatively specific and detailed, they should not be construed as limiting the scope of the present invention. It should be noted that a person skilled in the art could make various modifications and improvements without departing from the spirit of the present application, all of which fall within the scope of protection of the present application. Therefore, the scope of protection of the present patent application shall be determined by the appended claims.

Claims

1. A construction drawing review optimization method based on deep learning, characterized in that: The following steps are involved: Step 1: Obtain several construction drawing samples and annotate the construction drawing samples to obtain annotated samples; Step 2: Construct a deep learning model, and input the labeled samples into the deep learning model for training to construct a deep learning optimization model; Step 3: Obtain the construction drawing to be processed, and preprocess the construction drawing to be processed to obtain preprocessed data, and use the deep learning optimization model to identify and analyze the preprocessed data to extract key information from the preprocessed data; Step 4: Match and verify the key information with the construction drawing review specifications, and generate corresponding optimization strategies based on the matching verification results.

2. The construction drawing review optimization method based on deep learning according to claim 1 is characterized in that: The step one comprises: Collect construction drawing samples from various construction projects and obtain element types and corresponding positioning information of each construction drawing sample; Data annotation is performed on each of the construction drawing samples according to the element type and the positioning information to obtain a corresponding annotated sample.

3. The construction drawing review optimization method based on deep learning according to claim 1 is characterized in that: The step three includes: Obtaining the construction drawings to be processed uploaded by the user, and converting the format of the construction drawings to be processed to obtain preliminary construction drawings; The preliminary construction drawing is resized to obtain corresponding preprocessed data, and the image classification algorithm and target detection algorithm of the deep learning optimization model are used to perform image classification and analysis detection on the preprocessed data to extract corresponding key information.

4. The construction drawing review optimization method based on deep learning according to claim 1 is characterized in that: The fourth step includes: Match and verify the key information with the construction drawing review specifications, and mark data that does not comply with the construction drawing review specifications; A corresponding optimization strategy is generated from a preset strategy library based on the matching verification result, and the optimization strategy is fed back to the reviewer.

5. The construction drawing review optimization method based on deep learning according to claim 1 is characterized in that: The method further comprises: Feedback opinions are collected at preset intervals to form a corresponding feedback opinion database; The deep learning optimization model is optimized and improved according to the feedback library to obtain a deep learning improved model.

6. A construction drawing review optimization system based on deep learning, characterized in that: include: A data annotation module is used to obtain a number of construction drawing samples and perform data annotation on the construction drawing samples to obtain annotated samples; A model training module is used to build a deep learning model and input the labeled samples into the deep learning model for training to build a deep learning optimization model; A data processing module is used to obtain the construction drawings to be processed, pre-process the construction drawings to be processed to obtain pre-processed data, and use the deep learning optimization model to identify and analyze the pre-processed data to extract key information from the pre-processed data; The strategy generation module is used to match and verify the key information with the construction drawing review specifications, and generate corresponding optimization strategies based on the matching verification results.

7. The construction drawing review and optimization system based on deep learning according to claim 6 is characterized in that: The data annotation module includes: A data acquisition unit is used to collect construction drawing samples from various construction projects and obtain the element type and corresponding positioning information of each construction drawing sample; The data annotation unit is used to perform data annotation on each of the construction drawing samples according to the element type and the positioning information to obtain a corresponding annotation sample.

8. The construction drawing review and optimization system based on deep learning according to claim 6 is characterized in that: The data processing module includes: A format conversion unit, configured to obtain a construction drawing to be processed uploaded by a user, and convert the format of the construction drawing to be processed to obtain a preliminary construction drawing; A data processing unit is used to resize the preliminary construction drawing to obtain corresponding preprocessed data, and use the image classification algorithm and target detection algorithm of the deep learning optimization model to perform image classification and analysis and detection on the preprocessed data to extract corresponding key information.

9. The construction drawing review and optimization system based on deep learning according to claim 6 is characterized in that: The strategy generation module includes: A matching and verification unit, configured to match and verify the key information with the construction drawing review specification, and mark data that does not comply with the construction drawing review specification; The strategy generation unit is used to generate a corresponding optimization strategy from a preset strategy library according to the matching verification result, and feed back the optimization strategy to the reviewer.

10. The construction drawing review and optimization system based on deep learning according to claim 6 is characterized in that: The system further comprises: The feedback collection module is used to collect feedback at preset intervals to form a corresponding feedback database; The optimization and improvement module is used to optimize and improve the deep learning optimization model according to the feedback library to obtain a deep learning improved model.

Images