Method and device for intelligently identifying CAD drawing space through algorithm

Abstract

The invention belongs to the field of intelligent recognition of CAD drawing space, and particularly relates to a method and device for intelligently recognizing the CAD drawing space through an algorithm, and the method comprises the following steps: 1-A: inputting data of a functional room name library; 1-B, automatically recording the input data into a functional house name database; 1-1, an AutoCAD drawing is imported; 1-2, analyzing CAD drawing data; 1-3: obtaining a functional house library name list of the database; 1-4, matching house name labeling list data in the drawing data; 1-5, planning and identifying a spatial range by using a path planning algorithm; 1-6, identifying a part of space range by using a space enclosure algorithm; 1-7, non-repetitive space default reservation and repetitive space preferential reservation; according to the method, the identification efficiency is improved in magnitude, the identification precision is remarkably improved, the technical dependence and the labor cost are reduced, the method is deeply adaptive to a digital scene, and the method has strong scene adaptability.

Description

Technical Field

[0001] This invention relates to the field of intelligent recognition technology for CAD drawing space, and in particular to a method and apparatus for intelligent recognition of CAD drawing space through algorithms. Background Technology

[0002] In fields such as architectural design, engineering construction, and interior decoration, CAD drawings are the core carrier of spatial information. Traditional spatial identification in CAD drawings relies primarily on manual operation. Designers must analyze information such as room name markings, wall layouts, and door and window locations on each drawing to manually delineate spatial boundaries. This method has the following significant drawbacks:

[0003] Extremely inefficient: When faced with complex CAD drawings for large-scale building projects (such as drawings containing dozens or even hundreds of functional rooms), manually identifying the spatial scope takes a lot of time, which seriously slows down the project progress.

[0004] Poor accuracy: Human judgment is easily affected by subjective factors and visual fatigue, especially when dealing with irregular spaces and multiple complex spaces, which can easily lead to problems such as incorrect space division and deviation in range definition;

[0005] High technical threshold: It requires operators to have in-depth knowledge of architectural drafting and rich practical experience. The training period for newcomers is long and it is difficult to ensure the uniformity of recognition standards.

[0006] Unable to adapt to digital needs: The results of manual identification are mostly in the form of paper documents or scattered data, which are difficult to connect with digital systems such as BIM modeling, engineering cost estimation, and smart decoration, thus restricting the intelligent upgrading of the construction industry.

[0007] Although some CAD drawing parsing tools have emerged in the existing technology, most of them can only achieve basic element extraction and lack dedicated algorithms for "spatial intelligent recognition". In particular, there are technical bottlenecks in the association and matching of building name labels with spatial ranges and the automatic division of complex spaces, which cannot meet the needs of actual engineering for efficient and accurate spatial recognition.

[0008] Therefore, we propose a method and apparatus for intelligent spatial recognition of CAD drawings using algorithms to solve the above problems. Summary of the Invention

[0009] The purpose of this invention is to address the shortcomings of existing technologies by proposing a method and apparatus for intelligent spatial recognition of CAD drawings through algorithms.

[0010] To achieve the above objectives, the present invention adopts the following technical solution:

[0011] A method for intelligent spatial recognition of CAD drawings using algorithms includes the following steps:

[0012] 1-A: Data entry for the functional room name database;

[0013] 1-B: The entered data is automatically recorded in the functional room name database;

[0014] 1-1: Import AutoCAD drawings;

[0015] 1-2: Parsing CAD drawing data;

[0016] 1-3: Obtain the list of functional room names from the database;

[0017] 1-4: Match the list of house name labels in the drawing data;

[0018] 1-5: Use path planning algorithms to plan and identify the spatial range;

[0019] 1-6: Use spatial enclosure algorithms to identify a portion of the spatial range;

[0020] 1-7: Non-repeating spaces are reserved by default, while repeating spaces are reserved selectively.

[0021] 1-8: Output spatial range data.

[0022] Preferably, in steps 1-5, the path planning algorithm includes the following steps:

[0023] 2-A: Retrieves data parsed from CAD drawings;

[0024] 2-1: Calling the algorithm;

[0025] 2-2: Obtain the list of house name label data;

[0026] 2-3: Obtain a house name label from the house name label data list;

[0027] 2-4: Obtain the content, coordinates, and spatial range of a room name label;

[0028] 2-5: Filter nearby doors and windows within a certain range;

[0029] 2-6: Diffuse lines are emitted from the center point of the room name label and intersect with the rays from the center points of nearby doors and windows to obtain the shortest path and identify the nearest door or window;

[0030] 2-7: Determine if a door or window has been found. If yes, proceed to 2-8; otherwise, proceed to 2-9.

[0031] 2-8: Starting from the nearest door or window, and based on the solid line segments of the door, window, wall and column, try to plan a closed spatial path in both clockwise and counterclockwise directions. If the planned path is successful, it is identified as a spatial range.

[0032] 2-9: Determine if there is a next room name marked. If yes, jump to 2-3; otherwise, jump to 2-10.

[0033] 2-10: Output spatial range data.

[0034] Preferably, the spatial concealment algorithm includes the following steps:

[0035] 3-A: Retrieves data parsed from CAD drawings;

[0036] 3-1: Call the algorithm;

[0037] 3-2: Obtain the list of house name label data;

[0038] 3-3: Obtain a house name label from the house name label data list;

[0039] 3-4: Obtain the content, coordinates, and spatial range data of the room name label;

[0040] 3-5: Using the center point marked with the house name as the center, determine an initial range and filter the door, window, wall and column entities within the range;

[0041] 3-6: Use the selected entities as boundaries to enclose spaces and obtain all successfully enclosed spaces;

[0042] 3-7: Obtain the minimum enclosed space, including the house name label;

[0043] 3-8: Determine if the masking space has been obtained. If yes, jump to 3-11; otherwise, jump to 3-9.

[0044] 3-9: Determine if the maximum number of attempts to expand the space has been reached. If yes, jump to 3-11; otherwise, jump to 3-10.

[0045] 3-10: Expand the scope centered on the center point marked by the house name, re-filter the door, window, wall and column entities within the scope, and jump to 3-6;

[0046] 3-11: Determine if there is a next room name marked. If yes, jump to 3-3; otherwise, jump to 3-12.

[0047] 3-12: Output spatial range data.

[0048] Preferably, in steps 1-2, the open-source tools Teigha SDK API or LibreDWG are used to parse the drawing data.

[0049] Preferably, in steps 1-4, the room name label list is matched with the room name list from the drawing data, and the label content, coordinate position, and occupied space range data are saved.

[0050] Preferably, in steps 1-5, entities in the drawing that are not used for path planning are excluded based on the layers, contour features, and distance relationships of the entity data in the drawing. Based on the entity line segments of doors, windows, walls, and columns, the room name label is used as the algorithm entry point, and the spatial range is planned and identified using a spatial path planning algorithm.

[0051] Preferably, in steps 1-6, entities in the drawing that are not used for path planning are excluded based on the layers, contour features, and distance relationships of the entity data in the drawing. Based on the entity line segments of doors, windows, walls, and columns, the room name label is used as the algorithm entry point, and the spatial range is identified using a spatial enclosure algorithm.

[0052] Preferably, in steps 1-7, based on the spatial relationship of the house name labels, the spaces identified by the two algorithms are compared, and the space ranges that are repeatedly identified are selected and retained based on their area and outer contour data, while the non-repeating spaces are retained by default.

[0053] This invention also proposes a device for intelligent spatial recognition of CAD drawings through an algorithm, comprising:

[0054] The data entry module is used for entering data into the functional room name database;

[0055] The automatic recording module is used to automatically record the entered data into the functional room name database;

[0056] The import module is used to import AutoCAD drawings;

[0057] The parsing module is used to parse CAD drawing data;

[0058] The acquisition module is used to retrieve a list of functional room names from the database.

[0059] The matching module is used to match the list of house name labels in the drawing data;

[0060] The path planning algorithm module is used to plan and identify the spatial range using path planning algorithms.

[0061] The spatial concealment algorithm module is used to identify a portion of a spatial area using spatial concealment algorithms.

[0062] The comparison module is used to retain non-repeating spaces by default and select the best to retain repeating spaces.

[0063] The output module is used to output spatial range data.

[0064] Preferably, the import module, parsing module, acquisition module, matching module, path planning algorithm module, spatial enclosure algorithm module, comparison module, and output module are connected in sequence, and the data entry module, automatic recording module, and acquisition module are connected in sequence.

[0065] The beneficial effects of the method and apparatus for intelligent spatial recognition of CAD drawings using algorithms described in this invention are as follows:

[0066] Significantly improved recognition efficiency: The entire process of spatial recognition in CAD drawings is automated, and room name matching, space division, and range output can be completed without manual intervention. The processing efficiency is more than 10 times higher than that of manual operation, and the drawing parsing cycle is greatly shortened.

[0067] Significantly improved recognition accuracy: Through dual algorithm verification of "path planning + spatial enclosure" and combined with the comparison and selection of spatial relationships of house name labels, the spatial recognition accuracy can reach more than 98%, effectively avoiding errors from manual operation and ensuring the accuracy of spatial range definition;

[0068] Reduced reliance on technology and labor costs: No need for operators to have professional architectural drafting experience, the system can automatically complete the identification of complex spaces, lowering the technical threshold of the industry, while reducing manual input and saving labor costs;

[0069] Deeply adapted to digital scenarios: The output structured spatial range data can be directly connected to BIM modeling, engineering cost estimation, intelligent space planning and other systems, providing key data support for the digital and intelligent transformation of the construction industry;

[0070] It has strong scene adaptability: by dynamically adjusting the range parameters of spatial enclosure and optimizing the judgment logic of path planning, it can adapt to CAD drawings of different types and complexities (such as residential buildings, commercial complexes, industrial plants, etc.), solving the problem of insufficient recognition ability of existing technologies for complex spaces.

[0071] This invention offers an order-of-magnitude improvement in recognition efficiency, a significant increase in recognition accuracy, reduced reliance on technology and labor costs, deep adaptation to digital scenarios, and strong scenario adaptability. Attached Figure Description

[0072] Figure 1 This is a flowchart of the spatial recognition algorithm for a method of intelligent spatial recognition of CAD drawings proposed in this invention;

[0073] Figure 2 This is a flowchart of a spatial path planning algorithm for a method of intelligent spatial recognition of CAD drawings proposed in this invention;

[0074] Figure 3 This is a flowchart of a spatial enclosure algorithm for a method of intelligent spatial recognition of CAD drawings proposed in this invention. Detailed Implementation

[0075] The technical solutions of the present invention will be clearly and completely described below with reference to the accompanying drawings in the embodiments of the present invention. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments.

[0076] Example

[0077] Reference Figures 1-3 A method for intelligent spatial recognition of CAD drawings using algorithms includes the following steps:

[0078] 1-A: Data entry for the functional room name database;

[0079] 1-B: The entered data is automatically recorded in the functional room name database;

[0080] 1-1: Import AutoCAD drawings;

[0081] 1-2: Parsing CAD drawing data;

[0082] 1-3: Obtain the list of functional room names from the database;

[0083] 1-4: Match the list of house name labels in the drawing data;

[0084] 1-5: Use path planning algorithms to plan and identify the spatial range;

[0085] 1-6: Use spatial enclosure algorithms to identify a portion of the spatial range;

[0086] 1-7: Non-repeating spaces are reserved by default, while repeating spaces are reserved selectively.

[0087] 1-8: Output spatial range data.

[0088] In this embodiment, steps 1-5, the path planning algorithm includes the following steps:

[0089] 2-A: Retrieves data parsed from CAD drawings;

[0090] 2-1: Calling the algorithm;

[0091] 2-2: Obtain the list of house name label data;

[0092] 2-3: Obtain a house name label from the house name label data list;

[0093] 2-4: Obtain the content, coordinates, and spatial range of a room name label;

[0094] 2-5: Filter nearby doors and windows within a certain range;

[0095] 2-6: Diffuse lines are emitted from the center point of the room name label and intersect with the rays from the center points of nearby doors and windows to obtain the shortest path and identify the nearest door or window;

[0096] 2-7: Determine if a door or window has been found. If yes, proceed to 2-8; otherwise, proceed to 2-9.

[0097] 2-8: Starting from the nearest door or window, and based on the solid line segments of the door, window, wall and column, try to plan a closed spatial path in both clockwise and counterclockwise directions. If the planned path is successful, it is identified as a spatial range.

[0098] 2-9: Determine if there is a next room name marked. If yes, jump to 2-3; otherwise, jump to 2-10.

[0099] 2-10: Output spatial range data.

[0100] In this embodiment, the spatial concealment algorithm includes the following steps:

[0101] 3-A: Retrieves data parsed from CAD drawings;

[0102] 3-1: Call the algorithm;

[0103] 3-2: Obtain the list of house name label data;

[0104] 3-3: Obtain a house name label from the house name label data list;

[0105] 3-4: Obtain the content, coordinates, and spatial range data of the room name label;

[0106] 3-5: Using the center point marked with the house name as the center, determine an initial range and filter the door, window, wall and column entities within the range;

[0107] 3-6: Use the selected entities as boundaries to enclose spaces and obtain all successfully enclosed spaces;

[0108] 3-7: Obtain the minimum enclosed space, including the house name label;

[0109] 3-8: Determine if the masking space has been obtained. If yes, jump to 3-11; otherwise, jump to 3-9.

[0110] 3-9: Determine if the maximum number of attempts to expand the space has been reached. If yes, jump to 3-11; otherwise, jump to 3-10.

[0111] 3-10: Expand the scope centered on the center point marked by the house name, re-filter the door, window, wall and column entities within the scope, and jump to 3-6;

[0112] 3-11: Determine if there is a next room name marked. If yes, jump to 3-3; otherwise, jump to 3-12.

[0113] 3-12: Output spatial range data.

[0114] In this embodiment, steps 1-2 involve using open-source tools such as Teigha SDK API or LibreDWG to parse the drawing data.

[0115] In this embodiment, in steps 1-4, the room name label list is matched with the room name list from the drawing data, and the label content, coordinate position, and occupied space range data are saved.

[0116] In this embodiment, in steps 1-5, entities in the drawing that are not used for path planning are excluded based on the layer, contour features, and distance relationships of the entity data in the drawing. Based on the entity line segments of doors, windows, walls, and columns, the room name label is used as the algorithm entry point, and the spatial range is planned and identified using a spatial path planning algorithm.

[0117] In this embodiment, in steps 1-6, entities in the drawing that are not used for path planning are excluded based on the layer, contour features, and distance relationships of the entity data in the drawing. Based on the entity line segments of doors, windows, walls, and columns, the room name label is used as the algorithm entry point, and the spatial range is identified using the spatial enclosure algorithm.

[0118] In this embodiment, in steps 1-7, based on the spatial relationship of the house name labels, the spaces identified by the two algorithms are compared, and the space ranges that are repeatedly identified are selected and retained based on their area and outer contour data, while the non-repeating spaces are retained by default.

[0119] This embodiment also proposes a device for intelligent recognition of CAD drawing space through an algorithm, comprising:

[0120] The data entry module is used for entering data into the functional room name database;

[0121] The automatic recording module is used to automatically record the entered data into the functional room name database;

[0122] The import module is used to import AutoCAD drawings;

[0123] The parsing module is used to parse CAD drawing data;

[0124] The acquisition module is used to retrieve a list of functional room names from the database.

[0125] The matching module is used to match the list of house name labels in the drawing data;

[0126] The path planning algorithm module is used to plan and identify the spatial range using path planning algorithms.

[0127] The spatial concealment algorithm module is used to identify a portion of a spatial area using spatial concealment algorithms.

[0128] The comparison module is used to retain non-repeating spaces by default and select the best to retain repeating spaces.

[0129] The output module is used to output spatial range data.

[0130] In this embodiment, the import module, parsing module, acquisition module, matching module, path planning algorithm module, spatial enclosure algorithm module, comparison module, and output module are connected in sequence, and the data entry module, automatic recording module, and acquisition module are connected in sequence.

[0131] Reference Figure 1 Spatial recognition algorithm:

[0132] An expandable functional room library is built. The room names in the library are matched with the room names in the drawings. Calculations and judgments are made by combining path planning, spatial enclosure, graphic Boolean operations, topological relationships, directional relationships, and measurement relationships to plan or enclose a precise spatial range, thereby accurately assigning the engineering quantities in the drawings to their respective spaces.

[0133] Spatial recognition algorithm operation process description:

[0134] Process 1-A: Manually enter the functional room name database;

[0135] Process 1-B: The entered data is automatically recorded in the functional room name database;

[0136] Process 1-1: The user imports DWG format drawings from AutoCAD.

[0137] Process 1-2: Use open-source tools Teigha SDK API or LibreDWG to parse the drawing data.

[0138] Process 1-3: Retrieve the list of functional room names from the database.

[0139] Process 1-4: From the drawing data, match the list of house names with the list of house name annotations, and save the annotation content, coordinates, and space occupied by the annotations.

[0140] Process 1-5: Based on the layers, contour features, distance relationships, etc. of the entity data in the drawings, exclude entities in the drawings that are not used for path planning, and use the building name label as the algorithm entry point based on the entity line segments of doors, windows, walls and columns, and use the spatial path planning algorithm to plan and identify the spatial range.

[0141] Process 1-6: Based on the layers, contour features, distance relationships, etc. of the entity data in the drawings, exclude entities in the drawings that are not used for path planning, and use the building name label as the algorithm entry point based on the entity line segments of doors, windows, walls and columns, and use the spatial enclosure algorithm to identify the spatial range.

[0142] Process 1-7: Based on the spatial relationship marked by the house name, compare the spaces identified by the two algorithms, select the best space ranges that are repeatedly identified based on their area, outer contour and other data, and retain the non-repeating spaces by default.

[0143] Processes 1-8: Output spatial range data for use in front-end, back-end, and subsequent data recognition processes.

[0144] Detailed algorithm flow description:

[0145] Reference Figure 2 Spatial path planning algorithm:

[0146] Spatial path planning algorithm flow description:

[0147] Process 2-A: Call the data parsed from the CAD drawing.

[0148] Process 2-1: Process algorithm call.

[0149] Process 2-2: Obtain the list of building name annotation data from the drawing data parsed from the CAD drawings.

[0150] Step 2-3: Obtain a house name label from the house name label data list.

[0151] Steps 2-4: Obtain data such as the content of the house name label, coordinate location, and the space occupied.

[0152] Steps 2-5: Based on factors such as text size and image frame size, filter all nearby doors and windows within a certain range.

[0153] Process 2-6: Derive scattering lines from the center point of the house name label and intersect the rays emanating from the center points of nearby doors and windows to obtain the shortest path, so as to identify and obtain the nearest door or window and its physical data.

[0154] Step 2-7: Determine whether the door and window have been successfully acquired. If yes, proceed to Step 2-8; otherwise, proceed to Step 2-9.

[0155] Process 2-8: Starting from the nearest door or window entity, and based on the entity line segments of the door, window, wall and column, attempt to plan a closed spatial path in both clockwise and counterclockwise directions. If the planned path is successful, it is identified as a spatial range.

[0156] Process 2-9: Determine if there is a next room name label. If yes, jump to process 2-3 to continue running the next spatial path planning process; otherwise, jump to process 2-10.

[0157] Process 2-10: Output spatial range data for use in the next spatial algorithm process.

[0158] Reference Figure 3 Spatial enclosure algorithm:

[0159] Spatial Concealment Algorithm Flow Description:

[0160] Process 3-A: Call the data parsed from the CAD drawing.

[0161] Process 3-1: Process Algorithm Call.

[0162] Process 3-2: Obtain the list of building name annotation data from the drawing data parsed from the CAD drawings.

[0163] Process 3-3: Obtain a house name label from the house name label data list.

[0164] Steps 3-4: Obtain data such as the content of the house name label, coordinate location, and the space occupied.

[0165] Step 3-5: Using the center point of the house name label as the center, and the text size and frame size as the standard, determine an initial range, and filter the door, window, wall and column entities within the range.

[0166] Process 3-6: Use the selected entities as boundaries to enclose space (close the lines) and obtain all successfully enclosed spaces.

[0167] Step 3-7: Based on the location and size of the house name, obtain the smallest enclosed space within it.

[0168] Step 3-8: Determine whether the enclosed space, including the house name, has been successfully obtained. If so, proceed to Step 3-11; otherwise, proceed to Step 3-9.

[0169] Step 3-9: Determine if the maximum number of attempts to expand the search for the enclosed space centered on the house name mark has been reached. If yes, proceed to Step 3-11; otherwise, proceed to Step 3-10.

[0170] Process 3-10: Expand the scope centered on the house name mark, re-filter the doors, windows, walls and columns within the scope, and jump to process 3-6.

[0171] Process 3-11: Determine if there is a next room name label. If yes, proceed to process 3-3 to continue running the next space enclosure process; otherwise, proceed to process 3-12.

[0172] Process 3-12: Output spatial range data for use in the next spatial algorithm process.

[0173] Processes 3-3 to 3-11 are performed in parallel using multiple threads to improve the recognition efficiency of the spatial enclosure algorithm.

[0174] The above description is only a preferred embodiment of the present invention, but the scope of protection of the present invention is not limited thereto. Any equivalent substitutions or modifications made by those skilled in the art within the scope of the technology disclosed in the present invention, based on the technical solution and inventive concept of the present invention, should be covered within the scope of protection of the present invention.

Claims

1. A method for intelligent spatial recognition of CAD drawings using algorithms, characterized in that, Includes the following steps: 1-A: Data entry for the functional room name database; 1-B: The entered data is automatically recorded in the functional room name database; 1-1: Import AutoCAD drawings; 1-2: Parsing CAD drawing data; 1-3: Obtain the list of functional room names from the database; 1-4: Match the list of house name labels in the drawing data; 1-5: Use path planning algorithms to plan and identify the spatial range; 1-6: Use spatial enclosure algorithms to identify a portion of the spatial range; 1-7: Non-repeating spaces are reserved by default, while repeating spaces are reserved selectively; 1-8: Output spatial range data.

2. The method for intelligent spatial recognition of CAD drawings using an algorithm according to claim 1, characterized in that, In steps 1-5, the path planning algorithm includes the following steps: 2-A: Retrieves data parsed from CAD drawings; 2-1: Calling the algorithm; 2-2: Obtain the list of house name label data; 2-3: Obtain a house name label from the house name label data list; 2-4: Obtain the content, coordinates, and spatial range of a room name label; 2-5: Filter nearby doors and windows within a certain range; 2-6: Diffuse lines are emitted from the center point of the room name label and intersect with the rays from the center points of nearby doors and windows to obtain the shortest path and identify the nearest door or window; 2-7: Determine if a door or window has been found. If yes, proceed to 2-8; otherwise, proceed to 2-9. 2-8: Starting from the nearest door or window, and based on the solid line segments of the door, window, wall and column, try to plan a closed spatial path in both clockwise and counterclockwise directions. If the planned path is successful, it is identified as a spatial range. 2-9: Determine if there is a next room name marked. If yes, jump to 2-3; otherwise, jump to 2-10. 2-10: Output spatial range data.

3. The method for intelligent spatial recognition of CAD drawings using an algorithm according to claim 1, characterized in that, The spatial concealment algorithm includes the following steps: 3-A: Retrieves data parsed from CAD drawings; 3-1: Call the algorithm; 3-2: Obtain the list of house name label data; 3-3: Obtain a house name label from the house name label data list; 3-4: Obtain the content, coordinates, and spatial range data of the room name label; 3-5: Using the center point marked with the house name as the center, determine an initial range and filter the door, window, wall and column entities within the range; 3-6: Use the selected entities as boundaries to enclose spaces and obtain all successfully enclosed spaces; 3-7: Obtain the minimum enclosed space, including the house name label; 3-8: Determine if the masking space has been obtained. If yes, jump to 3-11; otherwise, jump to 3-9. 3-9: Determine if the maximum number of attempts to expand the space has been reached. If yes, jump to 3-11; otherwise, jump to 3-10. 3-10: Expand the scope centered on the center point marked by the house name, re-filter the door, window, wall and column entities within the scope, and jump to 3-6; 3-11: Determine if there is a next room name marked. If yes, jump to 3-3; otherwise, jump to 3-12. 3-12: Output spatial range data.

4. The method for intelligent spatial recognition of CAD drawings using an algorithm according to claim 1, characterized in that, In sections 1-2, open-source tools such as Teigha SDK API or LibreDWG are used to parse drawing data.

5. The method for intelligent spatial recognition of CAD drawings using an algorithm according to claim 1, characterized in that, In steps 1-4, the room name label list is matched with the room name list from the drawing data, and the label content, coordinate position, and occupied space range data are saved.

6. The method for intelligent spatial recognition of CAD drawings using an algorithm according to claim 1, characterized in that, In steps 1-5, entities in the drawings that are not used for path planning are excluded based on the layers, contour features, and distance relationships of the entity data in the drawings. Based on the entity line segments of doors, windows, walls, and columns, the room name label is used as the algorithm entry point, and the spatial range is planned and identified using the spatial path planning algorithm.

7. The method for intelligent spatial recognition of CAD drawings using an algorithm according to claim 1, characterized in that, In steps 1-6, entities in the drawings that are not used for path planning are excluded based on the layers, contour features, and distance relationships of the entity data in the drawings. Based on the entity line segments of doors, windows, walls, and columns, the room name label is used as the algorithm entry point, and the spatial range is identified using the spatial enclosure algorithm.

8. The method for intelligent recognition of CAD drawing space using an algorithm according to claim 1, characterized in that, In steps 1-7, based on the spatial relationship of the house name labels, the spaces identified by the two algorithms are compared. The space ranges that are repeatedly identified are selected and retained based on their area and outer contour data, while the non-repeating spaces are retained by default.

9. An apparatus for intelligent recognition of CAD drawing space using an algorithm, employing the method for intelligent recognition of CAD drawing space using an algorithm as described in any one of claims 1-8, characterized in that, include: The data entry module is used for entering data into the functional room name database; The automatic recording module is used to automatically record the entered data into the functional room name database; The import module is used to import AutoCAD drawings; The parsing module is used to parse CAD drawing data; The acquisition module is used to retrieve a list of functional room names from the database. The matching module is used to match the list of house name labels in the drawing data; The path planning algorithm module is used to plan and identify the spatial range using path planning algorithms. The spatial concealment algorithm module is used to identify a portion of a spatial area using spatial concealment algorithms. The comparison module is used to retain non-repeating spaces by default and select the best to retain repeating spaces. The output module is used to output spatial range data.

10. The device for intelligent spatial recognition of CAD drawings through an algorithm according to claim 9, characterized in that, The import module, parsing module, acquisition module, matching module, path planning algorithm module, spatial enclosure algorithm module, comparison module, and output module are connected in sequence, and the data entry module, automatic recording module, and acquisition module are connected in sequence.

Images