A method and system for identifying map sign information

Abstract

Embodiments of the present specification provide a kind of drawing signature information identification method, the method comprises: obtaining the drawing signature range of drawing;Based on the drawing frame of drawing and drawing signature range, determine the first closed path list;Based on the drawing element range of text drawing element in text drawing element list and the first closed path in first closed path list, determine the belonging closed path of drawing signature field;Based on the first intersection ratio of belonging closed path and drawing element range, and the offset distance of the range center point of text drawing element and belonging closed path, determine the content value range of drawing signature field;According to content value range, obtain the text content value in content value range.

Description

Technical Field

[0001] This manual relates to the field of construction technology, and in particular to a method and system for identifying drawing label information. Background Technology

[0002] In engineering construction and management, drawing management is a crucial link in ensuring the smooth progress of projects. With the development of digital technology, CAD electronic drawings have become an important tool for engineering design and management.

[0003] Although CAD software (such as AutoCAD and Tianzheng CAD) is widely used in drawing production, challenges remain in identifying title tag information. On one hand, title tag information for CAD electronic drawings is typically drawn by individual design institutes, resulting in variations in the naming and placement of title tag information, making it difficult to uniformly identify and manage. On the other hand, existing title tag information extraction methods mainly rely on manual annotation or simple geometric matching, which struggles to efficiently and accurately identify the range of content values ​​for title tag fields.

[0004] Currently, extracting drawing label information requires comprehensive consideration of the geometric features of the drawing elements (such as position, size, shape, etc.) and text content. Furthermore, when dealing with the drawing label standards of different design institutes, it is difficult to adapt to the diverse drawing label layouts and naming rules, resulting in low recognition efficiency and affecting the overall effectiveness of drawing management.

[0005] Therefore, it is necessary to provide a method and system for identifying drawing signature information, which can achieve efficient and accurate identification of drawing signature information, adapt to the drawing signature standards of different design institutes, and improve the overall efficiency of drawing management. Summary of the Invention

[0006] One embodiment of this specification provides a method for identifying title information. The method is executed by a processor and includes: obtaining the title range of a drawing; determining a first closed path list based on the drawing frame and the title range, the first closed path list including all first closed paths that do not contain other closed paths; determining the closed path to which the title field belongs based on the element range of text elements in a text element list and the first closed path in the first closed path list; determining the content value range of the title field based on a first intersection ratio between the belonging closed path and the element range, and the offset distance between the center point of the text element range and the belonging closed path; and obtaining the text content value within the content value range according to the content value range.

[0007] One embodiment of this specification provides a label information recognition system, characterized in that it includes: an acquisition module configured to acquire the label range of a drawing; a first determination module configured to determine a first closed path list based on the drawing frame and the label range, the first closed path list including all first closed paths that do not contain other closed paths; a second determination module configured to determine the closed path to which the label field belongs based on the element range of text elements in a text element list and the first closed path in the first closed path list; a third determination module configured to determine the content value range of the label field based on a first intersection ratio between the belonging closed path and the element range, and the offset distance between the center point of the text element range and the belonging closed path; and an extraction module configured to acquire text content values ​​within the content value range according to the content value range.

[0008] Some embodiments of this specification include at least the following beneficial effects: determining a first closed path based on the scope of the title block, determining the corresponding closed path, determining the content value range by using the intersection ratio between the corresponding closed path and the scope of the text element, and the offset distance between the center point of the text element's scope and the center point of the corresponding closed path, thereby extracting the text content value, which is beneficial for accurately determining the scope of title block fields in drawings of different design institutes and different standards, and improving the accuracy of identifying text content values. Attached Figure Description

[0009] This specification will be further described by way of exemplary embodiments, which will be described in detail with reference to the accompanying drawings. These embodiments are not limiting; in these embodiments, the same reference numerals denote the same structures, wherein:

[0010] Figure 1 These are exemplary block diagrams of a tag information recognition system according to some embodiments of this specification;

[0011] Figure 2 This is an exemplary flowchart of a label information recognition method according to some embodiments of this specification;

[0012] Figure 3 This is an exemplary schematic diagram illustrating the determination of a range of content values ​​based on a second closed path list according to some embodiments of this description;

[0013] Figure 4 This is an exemplary flowchart illustrating the determination of the scope of a title block according to some embodiments of this specification;

[0014] Figure 5 These are exemplary schematic diagrams of the labels shown according to some embodiments of this specification. Detailed Implementation

[0015] To more clearly illustrate the technical solutions of the embodiments in this specification, the accompanying drawings used in the description of the embodiments will be briefly introduced below. Obviously, the drawings described below are merely some examples or embodiments of this specification. For those skilled in the art, these drawings can be applied to other similar scenarios without creative effort. Unless obvious from the context or otherwise specified, the same reference numerals in the drawings represent the same structures or operations.

[0016] It should be understood that the terms "system," "unit," and / or "module" used herein are a method of distinguishing different components, elements, parts, sections, or assemblies at different levels. However, if other terms can achieve the same purpose, they may be replaced by other expressions.

[0017] As indicated in this specification and claims, unless the context clearly indicates otherwise, the words "a," "an," "an," and / or "the" do not specifically refer to the singular and may also include the plural. Generally speaking, the terms "comprising" and "including" only indicate the inclusion of expressly identified steps and elements, which do not constitute an exclusive list, and the method or apparatus may also include other steps or elements.

[0018] Flowcharts are used in this specification to illustrate the operations performed by the system according to embodiments of this specification. It should be understood that the preceding or following operations are not necessarily performed in exact order. Instead, the steps can be processed in reverse order or simultaneously. Furthermore, other operations can be added to these processes, or one or more steps can be removed from them.

[0019] Figure 1 These are exemplary block diagrams of a label information recognition system according to some embodiments of this specification. It should be noted that the following embodiments are for illustrative purposes only and do not constitute a limitation thereof.

[0020] In some embodiments, such as Figure 1 As shown, the image tag information recognition system 100 may include an acquisition module 110, a first determination module 120, a second determination module 130, a third determination module 140, and an extraction module 150. A processor may be integrated into one or more of these modules.

[0021] In some embodiments, one or more modules in the image tag information recognition system 100 may exchange information and / or data via a network. In some embodiments, the network may be any one or more of a wired network or a wireless network.

[0022] In some embodiments, the acquisition module 110 can be configured to acquire the label range of a drawing.

[0023] In some embodiments, the acquisition module 110 may also be configured to generate a list of graphic element information based on text graphic elements in the drawing; determine an aggregation range list based on the aggregation of text graphic elements in the graphic element information list, wherein the aggregation range list includes the graphic element range of at least one text graphic element; and determine the label range based on the aggregation range list.

[0024] In some embodiments, the acquisition module 110 may also be configured to, for each text element in the element information list, calculate a first intersection-union ratio (IUGR) between the element range of the text element and the frame range of the frame; delete the text element from the element information list in response to the first IUGR being greater than a third threshold; fill the element range of the text element into the aggregation range list in response to the first IUGR being less than or equal to the third threshold and in response to the aggregation range list being empty; determine the union of each aggregation range in the aggregation range list with the element range of the text element in response to the aggregation range list not being empty, and calculate a second IUGR between the union and the frame; fill the element range of the text element into the aggregation range list in response to the second IUGR being greater than a fourth threshold; replace the aggregation range with the union in response to the second IUGR being less than or equal to the fourth threshold; and determine the tag range based on the number of occurrences of the tag field in at least one element range in the aggregation range list.

[0025] In some embodiments, the first determining module 120 may be configured to determine a first closed path list based on the drawing frame and title block range of the drawing, the first closed path list including all first closed paths that do not contain other closed paths.

[0026] In some embodiments, the first determining module 120 may also be configured to determine the line segments that intersect the frame and the label range; construct an undirected graph based on the line segments and the intersection of the frame and the label range; traverse the undirected graph to determine multiple third closed paths; and determine a first closed path list based on the inclusion relationship between the multiple third closed paths.

[0027] In some embodiments, the second determining module 130 may be configured to determine the closed path to which the label field belongs based on the element range of the text elements in the text element list and the first closed path in the first closed path list.

[0028] In some embodiments, the second determining module 130 may also be configured to generate a list of graphic elements based on text graphic elements in the drawing; and to generate a list of text graphic elements based on the list of graphic elements and a common text information table.

[0029] In some embodiments, the third determining module 140 may be configured to determine the content value range of the label field based on a first intersection ratio between the belonging closed path and the primitive range, and the offset distance between the center point of the text primitive range and the center point of the belonging closed path.

[0030] In some embodiments, the third determining module 140 may also be configured to determine whether the first intersection ratio and offset distance meet the first preset condition; and in response to the first intersection ratio and offset distance meeting the first preset condition, to take the range of the closed path to which it belongs as the range of content values.

[0031] In some embodiments, the third determining module 140 may also be configured to, in response to the first intersection ratio and offset distance not meeting the first preset condition, determine adjacent closed paths sharing vertices with the belonging closed path; determine adjacent closed paths that meet the filtering conditions as second closed paths and generate a second closed path list; and determine the content value range based on the second closed path list.

[0032] In some embodiments, the third determining module 140 may further be configured to: determine the number of paths in the second closed path list; determine the corresponding closed path as a content value range in response to the number of paths in the second closed path list being 0; determine the second intersection ratio between the second closed path and the tag range in response to the number of paths in the second closed path list being greater than 0; determine the second intersection ratio between the second closed path and the tag range as a content value range in response to the number of second closed paths in the second closed path list satisfying the second preset condition being less than a preset number; and calculate the target distance in a preset direction between the center point of each second closed path and the center point of the corresponding closed path in the second closed path list, and determine the second closed path whose target distance in a preset direction between the center point of the corresponding closed path and the center point of the corresponding closed path is satisfied as a third preset condition, in response to the number of second closed paths in the second closed path list satisfying the second preset condition for the second intersection ratio of the tag range being not less than a preset number.

[0033] In some embodiments, the extraction module 150 can be configured to obtain text content values ​​within a content value range based on the content value range.

[0034] For more information on the above modules, please refer to [link / reference]. Figures 2-4 And related explanations.

[0035] It should be noted that the above description of the image tag information recognition system 100 and its modules is for convenience only and should not be construed as limiting this specification to the scope of the embodiments described. It is understood that those skilled in the art, after understanding the principles of the system, may arbitrarily combine the various modules or construct subsystems connected to other modules without departing from these principles. In some embodiments, Figure 1 The acquisition module 110, the first determination module 120, the second determination module 130, the third determination module 140, and the extraction module 150 disclosed herein can be different modules within a single system, or a single module can implement the functions of two or more of the aforementioned modules. For example, the modules can share a single storage module, or each module can have its own separate storage module. Such variations are all within the scope of protection of this specification.

[0036] Figure 2 This is an exemplary flowchart of a label information recognition method according to some embodiments of this specification.

[0037] In some embodiments, process 200 can be executed by the image label information recognition system 100. For example... Figure 2 As shown, process 200 includes the following steps:

[0038] Step 210: Obtain the label range of the drawing.

[0039] Drawings are diagrams used to guide the construction of a project. Drawings can include illustrative representations of details required for construction, such as dimensions, orientation, and technical parameters, as well as illustrative representations of the actual engineering structure to be constructed.

[0040] In some embodiments, drawings may include architectural construction drawings, structural construction drawings, and equipment construction drawings. Architectural construction drawings are primarily used to represent the planned location, external shape, layout of internal rooms, interior and exterior decoration, and construction requirements of a building. Structural construction drawings primarily represent the structural type, layout, component types, quantities, sizes, and construction methods of the building's load-bearing structure. Equipment construction drawings primarily represent the layout and construction requirements of the building's water supply and drainage, heating and ventilation, power supply, and lighting equipment.

[0041] In some embodiments, the drawings can be drawings created for the entire construction project. In some embodiments, the drawings can be drawings created for a specific construction object within the construction project. A construction project is a specific project for which construction or engineering work is to be carried out. Taking the construction project as building a school as an example, the drawings can be drawings created for the entire school. The drawings can also be drawings created for a specific teaching building. The drawings can also be drawings created for a specific floor of the teaching building.

[0042] In some embodiments, the drawings may be electronic drawings generated by CAD software, or drawings that are converted from paper drawings into electronic format through scanning or other means.

[0043] The label area refers to the area occupied by the label on the drawing. The label is an element used to identify information on the drawing. Drawing information refers to information related to the construction project contained on the drawing. For example, drawing information may include the drawing name, number, designer's name, design date, and company name. Alternatively, drawing information may include project information, building information, floor information, construction requirements, and drawing dimensions (e.g., length, width, and area). For instance, the label may include elements such as the drawing title, drawing type, drawing number, designer's name, design date, and company name.

[0044] The area of ​​the label can be a rectangular region. The extent of the rectangular region can be defined by the maximum and minimum x-coordinates, maximum and minimum y-coordinates of the vertices of the shape enclosed by the rectangular region.

[0045] In some embodiments, the processor can obtain the label range in various ways. For example, the processor can determine the area of ​​the smallest bounding rectangle that completely contains the label (e.g., drawing name, drawing type, drawing number, designer's name, design date, organization name, etc.) as the label range.

[0046] In some embodiments, the processor may also generate a list of graphic element information based on text graphic elements in the drawing; determine a list of aggregation ranges based on the aggregation of text graphic elements in the list of graphic element information; and determine the label range based on the list of aggregation ranges. Further description of this embodiment can be found in [reference needed]. Figure 4 And related content.

[0047] Step 220: Determine the first closed path list based on the drawing frame and title block range.

[0048] In engineering drawings, a drawing frame refers to the outline that defines the drawing area on a sheet of paper, usually drawn with a thick solid line. The drawing frame is used to indicate the drawing area of ​​a sheet.

[0049] The first closed path list consists of multiple first closed paths.

[0050] The first closed path refers to a closed path within the label area that does not contain any other closed paths, i.e., the smallest rectangle within the label area.

[0051] A closed path is a path that connects a starting point and an ending point. A path can be a line segment on a drawing. A closed path can form a closed shape. For example, in a tabular title block containing multiple closed rectangles, the smallest rectangle that does not contain any other rectangles is the first closed path.

[0052] In some embodiments, the processor can determine a first closed path based on a variety of methods, and then determine a list of first closed paths. For example, the processor can determine a first closed path in multiple map regions within a frame through image recognition (e.g., connected component analysis, Hough transform, etc.), and generate a list of first closed paths based on one or more first closed paths.

[0053] In some embodiments, the processor may determine the line segments that intersect the frame and the label range; construct an undirected graph based on the line segments and the intersection of the frame and the label range; traverse the undirected graph to determine multiple closed paths; and determine a first list of closed paths based on the inclusion relationship between the multiple closed paths.

[0054] In some embodiments, line segments in a drawing may include graphic outline lines, annotation line segments, dimension lines, etc.

[0055] In some embodiments, the processor can determine the straight line segments that intersect the drawing and the label area based on computer recognition algorithms (such as line intersection analysis).

[0056] An undirected graph is a mathematically structured graph consisting of vertices and edges. The vertices of an undirected graph are the intersections of line segments, and the edges are line segments. An edge is formed when the endpoints of two line segments corresponding to two intersection points are also endpoints of a line segment. The edges of an undirected graph have no direction.

[0057] In some embodiments, the processor can construct an undirected graph using various methods. For example, the processor can use a computer recognition algorithm to identify the line segments that intersect the drawing and the label area, and use the intersection points formed by the intersection of the line segments as vertices of the undirected graph; based on multiple vertices, it can determine whether any two vertices are the two endpoints of the same line segment, and if so, connect the two vertices to form an edge of the undirected graph to construct the undirected graph.

[0058] A closed path is a path that connects a starting point and an ending point, forming a closed shape. In some embodiments, multiple closed paths may have an inclusion or containment relationship.

[0059] In some embodiments, the processor may determine a closed path that does not contain other closed paths (i.e., the smallest closed path) as a first closed path and construct a first closed path list based on the first closed path.

[0060] In some embodiments of this specification, determining the first closed path list based on an undirected graph is beneficial for accurately determining closed paths that do not contain other closed paths, thereby making the first closed path list more accurate and complete.

[0061] Step 230: Based on the text element range in the text element list and the first closed path in the first closed path list, determine the closed path to which the tag field belongs.

[0062] A text element list is a list consisting of text elements from a drawing.

[0063] Text primitives refer to the physical entities containing text in a drawing. Text primitives can include text information (e.g., drawing name, drawing number). In some embodiments, text primitives may also contain other information related to the text, such as location points (e.g., coordinates (50, 120)) and extent information (e.g., [x...]). min =50,y min =100,x max =150,y max =120], where x min and x max Let y be the minimum and maximum x-coordinates of each vertex of the rectangle containing the text primitive (i.e., the area of ​​the text primitive). min and y max These are the minimum and maximum ordinates of each vertex of the rectangle containing the text element, respectively.

[0064] In some embodiments, the processor can acquire text elements in a drawing using computer recognition algorithms. These algorithms may include image recognition, OCR recognition, etc.

[0065] In some embodiments, the processor may obtain a list of text primitives based on user input.

[0066] In some embodiments, the processor can determine the text elements contained in the drawing, and then determine a list of text elements. For example, the processor can cluster the text elements determined from the drawing, and identify text elements belonging to the same cluster category as the same text element in the list of text elements. For example, texts such as "drawing name", "drawing title", "drawing name", and "drawing name" can be clustered into the same cluster category and identified as the same text element in the list of text elements.

[0067] The processor can determine text elements from a drawing in various ways. In some embodiments, the processor can acquire elements within a drawing frame based on parsing techniques. The drawing frame may contain multiple types of elements. The processor can further determine text elements from these multiple types of elements.

[0068] Analysis techniques refer to the techniques used to analyze drawings. Analysis techniques can include geometric analysis techniques, text analysis techniques, layer analysis techniques, attribute analysis techniques, etc.

[0069] Graphical elements refer to the basic elements in a drawing. Graphical elements include various types such as text elements, line segments, and polylines. A polyline can be broken down into multiple line segments. For example, a polyline can be a closed rectangle formed by multiple line segments, or a broken line formed by connecting multiple line segments.

[0070] In some embodiments, the processor may determine text primitives from multiple types of primitives based on computer recognition technology, etc.

[0071] In some embodiments, the processor can also generate a list of graphic elements based on text graphic elements in the drawing; and generate a list of text graphic elements based on the list of graphic elements and a table of commonly used text information.

[0072] The primitive information list includes one or more text primitives.

[0073] In some embodiments, the processor can fill all the text elements identified from the drawing into the element information list one by one.

[0074] In some embodiments, the list of commonly used text information includes multiple commonly used text information.

[0075] Commonly used text information refers to the frequently used textual descriptions of the title tag field or the text corresponding to the title tag field value in different drawings. For example, the title tag field "Drawing Name" is generally "Drawing Title" in Design Institute A's drawings, and generally "Drawing Name" in Design Institute B's drawings. Therefore, both "Drawing Title" and "Drawing Name" are commonly used textual information for the title tag field "Drawing Name". Title tag field values ​​refer to the content values ​​of the title tag field. Title tag fields are fields used to represent the attributes of the title tag. For example, title tag fields may include the title tag, drawing number, and drafter's name. Title tag field values ​​may include the content values ​​corresponding to the title tag, the drawing number, and the drafter's name, etc.

[0076] In some embodiments, the list of frequently used text information may be constructed by technicians based on experience. In some embodiments, the processor may construct the list of frequently used text information based on historical data. For example, the processor may analyze drawings from different design institutes, identify text content that appears frequently (e.g., exceeding a certain threshold) as frequently used text information, and add it to the list of frequently used text information.

[0077] In some embodiments, the list of commonly used text information includes multiple commonly used text information (i.e., the text of text primitives).

[0078] In some embodiments, the processor can acquire image tag information and commonly used text information; and determine a commonly used text information table based on the image tag information and commonly used text information.

[0079] In some embodiments, the label information includes a label field and a label field value.

[0080] In some embodiments, the processor can determine a common text information table based on the title block information, common text information, and the correspondence between the title block information and the common text information. For example, the processor determines the common text information of each title block information in historical drawings based on historical data. For each title block information, the processor can determine one or more historical drawings corresponding to the title block information. Each historical drawing may include a common text information corresponding to the title block information. The title block information and the common text information in the one or more historical drawings corresponding to it are constructed into a record, and a common text information table is constructed based on multiple records.

[0081] In some embodiments, the processor can generate a list of text elements in various ways. For example, the processor can traverse the element information list and determine whether a certain text element in the element information list is included in a common text information table (i.e., the text element has at least one field in the common text information table). If it is included, the processor records the tag field corresponding to the text element and adds the tag field to the text element list; if it is not included, the processor continues to traverse the next text element in the element information list.

[0082] In some embodiments of this specification, a text primitive list is generated based on a primitive information list and a commonly used text information list. This facilitates the more accurate generation of the text primitive list and reduces errors caused by manual generation.

[0083] The primitive extent refers to the bounding rectangle containing the primitive. For example, the primitive extent of a text primitive can be the rectangular area formed by the maximum and minimum ordinates, maximum and minimum abscissas of the bounding rectangle of the text primitive. As another example, for a line segment, the primitive extent of the line segment can be the coordinates of the endpoints of the line segment and the line connecting the two endpoints.

[0084] The closed path to which a title field belongs refers to the complete containment between the text element of the title field and a closed path that does not contain other closed paths.

[0085] In some embodiments, for any text element in the text element list, the element range processor can determine the inclusion relationship between the element range and each closed path that does not contain other closed paths, and determine the closed path that completely contains the element range and does not contain other closed paths as the closed path to which the tag field corresponding to the text element belongs.

[0086] Step 240: Based on the first intersection ratio between the closed path and the primitive range, and the offset distance between the center point of the text primitive range and the center point of the closed path, determine the content value range of the tag field.

[0087] The first intersection ratio is a metric used to measure the degree of overlap between the closed path to which a tag field belongs and the geographic extent of the corresponding text element. The first intersection ratio can be expressed numerically; the larger the value of the first intersection ratio, the greater the degree of overlap between the closed path and the geographic extent.

[0088] In some embodiments, the processor may determine the overlapping area between the closed path to which the tag field belongs and the primitive range of the text primitive corresponding to the tag field, and determine a first intersection ratio based on the overlapping area and the closed path. For example, the processor may determine the area of ​​the overlapping area between the primitive range of the text primitive and the closed path, and determine the first intersection ratio as the ratio of the area of ​​the overlapping area to the enclosed area corresponding to the closed path.

[0089] The center point of the range refers to the center point of the range of text primitives.

[0090] In some embodiments, the processor can determine the range center point based on the geometric center point of the text primitive's primitive range. For example, if the primitive range of the text primitive is a rectangular area, the processor can determine the geometric center point of that rectangular area as the range center point.

[0091] Offset distance refers to the distance in the target direction between the center point of the text element's range and the center point of the area enclosed by its closed path. The target direction can be either the horizontal or vertical axis of the drawing, and can be set according to actual needs.

[0092] In some embodiments, the processor can calculate the relative distance in the target direction between the center point of the text primitive range and the center point of the enclosed area of ​​the corresponding closed path, and thus determine the offset distance.

[0093] The content value range refers to the range of text elements corresponding to the tag field value of the tag field.

[0094] In some embodiments, the processor may base its work on a first intersection ratio between the closed path and the primitive's extent, and the offset distance between the center point of the text primitive's extent and the center point of its closed path.

[0095] For more information on how to determine the range of content values, please refer to [link / reference]. Figure 3 And its related descriptions.

[0096] Step 250: Obtain the text content value within the content value range based on the content value range.

[0097] The text content value refers to the text within the content value range. For example, if cell B in the content value range A contains the text "Electrical Design Specification", then "Electrical Design Specification" is the text content value of the content value range A.

[0098] In some embodiments, the processor may acquire text content values ​​within a specified range based on computer recognition algorithms. These algorithms may include image recognition, OCR recognition, etc.

[0099] In some embodiments of this specification, a first closed path is determined based on the scope of the title block, and the corresponding closed path is determined. The content value range is determined by the intersection ratio between the corresponding closed path and the scope of the text element, as well as the offset distance between the center point of the text element's scope and the center point of the corresponding closed path, thereby extracting the text content value. This is beneficial for accurately determining the scope of title block fields in drawings from different design institutes and according to different standards, and improves the accuracy of identifying text content values.

[0100] It should be noted that the above description of process 200 is for illustrative purposes only and does not limit the scope of this specification. Those skilled in the art can make various modifications and changes to process 200 under the guidance of this specification. However, these modifications and changes remain within the scope of this specification.

[0101] Figure 3 This is an exemplary schematic diagram illustrating the determination of a range of content values ​​according to some embodiments of this description.

[0102] like Figure 3 As shown, in some embodiments, the processor can also determine the content value range based on the closed path to which the tag field belongs, a first intersection ratio 310 between the tag field and the primitive range of the text primitive corresponding to the tag field, an offset distance 320 between the center point of the range of the text primitive corresponding to the tag field and the center point of the closed path to which the tag field belongs, and the closed path to which the tag field belongs.

[0103] like Figure 3As shown, in some embodiments, the processor can determine whether the first intersection ratio 310 and the offset distance 320 satisfy a first preset condition; and in response to the first intersection ratio 310 and the offset distance 320 satisfying the first preset condition, the range of the closed path to which they belong is taken as the range of content values.

[0104] In some embodiments, the first preset condition may include a weighted sum of the first intersection ratio and the offset distance that is greater than a preset threshold. The weights corresponding to the first intersection ratio and the offset distance, and the preset threshold, may be preset by the processor based on default settings or by a technician based on experience.

[0105] In some embodiments, the first preset condition may further include a first intersection ratio less than or equal to a first threshold, and the ratio of the offset distance to the target direction length of the closed path being to which it belongs is greater than a second threshold. Here, the target direction length refers to the projected length of the closed path in the target direction.

[0106] The first threshold refers to a threshold related to the first intersection ratio. For example, the first threshold can be any value between 0.1 and 0.6. Another example is that the first threshold can be any value between 0.2 and 0.5. Yet another example is that the first threshold can also be any value between 0.25 and 0.45. When the first intersection ratio is less than or equal to the first threshold, the degree of overlap between the closed path and the primitive range is relatively small.

[0107] In some embodiments, the first threshold may be a system default value, an empirical value, a manually preset value, or any combination thereof, and can be set according to actual needs.

[0108] The second threshold is a threshold related to the offset distance. For example, the second threshold can be any value between 0.1 and 0.7. Another example is that the second threshold can be any value between 0.2 and 0.6. Yet another example is that the second threshold can also be any value between 0.25 and 0.55. When the ratio of the offset distance to the target direction length of the corresponding closed path is greater than the second threshold, the text element corresponding to the tag field and the closed path to which the tag field belongs are within the same closed path.

[0109] In some embodiments, the second threshold may be a system default value, an empirical value, a manually preset value, or any combination thereof, and can be set according to actual needs.

[0110] In some embodiments of this specification, determining the first preset condition by using a first threshold and a second threshold is beneficial for reducing erroneous identification of content value ranges, improving the identification efficiency of content value ranges, and reducing redundant calculations.

[0111] The scope of a closed path refers to the area enclosed by the straight line segments that make up the closed path.

[0112] In some embodiments of this specification, determining the range of the corresponding closed path as the content value range based on judging whether the first intersection ratio and the offset distance satisfy the first preset condition is beneficial to accurately determining the content value range of title block fields in drawings prepared by different design institutes with different standards, thereby improving the accuracy of recognizing text content values.

[0113] See Figure 3 , in some embodiments, the processor may further, in response to the first intersection ratio 310 and the offset distance 320 not satisfying the first preset condition, determine an adjacent closed path 330 sharing a vertex with the corresponding closed path; determine the adjacent closed path satisfying the screening condition as a second closed path 340, and generate a second closed path list 350; and determine the content value range 360 based on the second closed path list 350.

[0114] In some embodiments, that the first intersection ratio and the offset distance do not satisfy the first preset condition includes that the first intersection ratio is not less than or equal to the first threshold, and / or the ratio of the offset distance to the target-direction length of the corresponding closed path is not greater than the second threshold.

[0115] Adjacent closed paths refer to closed paths that share a vertex with the corresponding closed path. For example, Figure 5 is an exemplary schematic diagram of a title block according to some embodiments of this specification. As shown in Figure 5 , the cell where "Drawn by" is located is the corresponding closed path of the title block field "Drawn by", and the adjacent closed paths of this corresponding closed path may include the closed paths where the text primitives "X University Architectural Design and Research Institute", "Personnel A", "Designed by", and "Personnel B" are located.

[0116] In some embodiments, if the first intersection ratio is less than or equal to the first threshold, and the ratio of the offset distance to the target-direction length of the corresponding closed path is greater than the second threshold, then the first intersection ratio and the offset distance satisfy the first preset condition. For example, as shown in Figure 5 , taking the title block field "Project Name" as an example, the first intersection ratio between the range of "Project Name" (that is, the primitive range of the text primitive corresponding to the text "Project Name") and the closed path to which "Project Name" belongs is between 0.1 and 0.5, and the projection distance between the center point of the primitive range of the title block field "Project Name" and the center point of the corresponding closed path in the target direction (that is, the horizontal direction in the figure) is greater than 0.1 to 0.7 times the projection length of the corresponding closed path in the horizontal direction, then it is considered that the first intersection ratio and the offset distance between the title block field "Project Name" and the corresponding closed path satisfy the first preset condition. Then it is considered that the content value "Jia" of the field and the text primitive "Project Name" are within the same closed path, so the range of the corresponding closed path is the range of the content value of the title block field, and the text content value can be determined therefrom, which is "Jia".

[0117] In some embodiments, if the first intersection ratio is not less than or equal to a first threshold, and / or the ratio of the offset distance to the target direction length of the corresponding closed path is not greater than a second threshold, then the first intersection ratio and / or the offset distance do not meet the first preset condition. For example, Figure 5 As shown, taking the tag field "cartography" as an example, if the distance between the center point of the primitive range of "cartography" (i.e. the primitive range of the text primitive corresponding to the text "cartography") and the center point of its corresponding closed path in the target direction (i.e. the horizontal direction in the figure) is greater than 0.7 times the horizontal projection length of the closed path, it is considered that the tag field "cartography" does not meet the first preset condition, and the adjacent closed paths that share a vertex with the closed path to which "cartography" belongs are further determined.

[0118] In some embodiments, the processor can determine adjacent closed paths using various methods. For example, the processor can determine adjacent closed paths that share vertices with other closed paths by traversing an undirected graph.

[0119] The second closed path list refers to a list containing second closed paths. A second closed path is an adjacent closed path that is not occupied by any title field in the drawing or is occupied by a text element corresponding to a title field value. In some embodiments, the second closed path list may be empty.

[0120] In some embodiments, the filtering criteria include that the second closed path is not occupied by any title field in the drawing or by the text element corresponding to the title field value. For example, if among the adjacent closed paths of a given closed path, there is an adjacent closed path that is neither determined to be the closed path of a title field nor the closed path of the text element corresponding to the title field value, then that adjacent closed path can be used as the second closed path. Figure 5 Among the adjacent closed paths "Personnel A", "Design", "X University Architectural Design Institute", and "Personnel B" that share a vertex with the closed path belonging to "Drafting", exclude closed paths that have already been occupied by the title field, such as "X University Architectural Design Institute" and "Design"; then exclude closed paths that have already been occupied by the content value of other fields, such as the closed path occupied by the content value "Personnel B" of the "Design" title, and take the closed path belonging to "Personnel A" as the second closed path.

[0121] In some embodiments, the processor may determine the range of content values ​​based on the number of second closed paths in the second closed path list.

[0122] In some embodiments, the processor may determine the number of paths in the second closed path list; in response to the number of paths in the second closed path list being 0, determine the corresponding closed path as a content value range; in response to the number of paths in the second closed path list being greater than 0, determine the aggregation range of each second closed path in the second closed path list with its corresponding path range, and determine the second intersection ratio of each aggregation range with the tag range; in response to the number of second closed paths whose second intersection ratio satisfies a second preset condition being less than a preset number, use the second closed paths whose second intersection ratio satisfies the second preset condition as content value ranges; and in response to the number of second closed paths whose second intersection ratio satisfies the second preset condition being not less than a preset number, calculate the target distance between the center point of each second closed path and the center point of its corresponding closed path in a preset direction, and use the second closed paths whose target distance with the center point of their corresponding closed path in the preset direction satisfies a third preset condition as content value ranges.

[0123] The number of paths refers to the number of second closed paths in the second closed path list.

[0124] The aggregation range can be the union of the second closed path and the path range to which the tag field belongs.

[0125] The second intersection ratio refers to the intersection ratio between the aggregation range and the label range. The second intersection ratio can be expressed numerically. The larger the second intersection ratio, the greater the intersection between the second closed path and the label range. In some embodiments, the processor can determine the second intersection ratio as the ratio between the area of ​​the aggregation range and the area of ​​the label range.

[0126] In some embodiments, the second preset condition includes a second intersection ratio being less than a fifth threshold. The fifth threshold can be represented by a numerical value. For example, the fifth threshold can be any value between 0.2 and 0.5.

[0127] In some embodiments, the second preset condition includes minimizing the second crossover ratio.

[0128] In some embodiments, the second preset condition may be a system default value, an empirical value, a manually preset value, or any combination thereof, and can be set according to actual needs.

[0129] The preset quantity can be expressed numerically. In some embodiments, the preset quantity can be a system default value, an empirical value, a manually preset value, or any combination thereof, and can be set according to actual needs. For example, the preset quantity can be 1.

[0130] The target distance refers to the distance between the center point of the second closed path and the center point of its parent closed path in a preset direction. In some embodiments, the preset direction may be the vertical axis of the drawing.

[0131] In some embodiments, the third preset condition includes minimizing the target distance.

[0132] In some embodiments, the processor may use a second closed path whose target distance satisfies a third preset condition as the content value range. For example, the processor may calculate the value of the center point of the second closed path and the center point of its parent closed path in the vertical axis direction, and use the second closed path corresponding to the center point of the second closed path with the smallest value in the vertical axis direction as the content value range.

[0133] As an example only, the processor can obtain the number of second closed paths contained in the second closed path list. When the number of paths is determined to be 0, the closed path belonging to the tag field is determined as the content value range. That is, when there are no adjacent closed paths, the content value range is further determined within the closed paths belonging to the tag field. When the number of second closed paths contained in the second closed path list is not 0, i.e., it contains multiple adjacent closed paths, the second intersection ratio of each second closed path in the second closed path list with the tag range is further determined, and the second closed path whose second intersection ratio is less than the fifth threshold is determined (hereinafter referred to as the target second closed path).

[0134] When the number of target second closed paths is less than the preset number, the target second closed paths are determined as the content value range; when the number of target second closed paths is not less than the preset number, the target distance between the center point of each target second closed path and the center point of its corresponding closed path in the vertical axis direction of the drawing is further determined, and the target second closed path with the smallest target distance is determined as the content value range.

[0135] In some embodiments of this specification, the content value range is determined based on the number of paths in the second closed path list, the second preset conditions, and the preset quantity. This is beneficial for dynamically adjusting the strategy for identifying the content value range, adapting to the drawing structures of different drawings from different design institutes, thereby more accurately determining the content value range and improving the accuracy of content value range identification.

[0136] In some embodiments of this specification, adjacent closed paths are determined based on a first intersection ratio, offset distance, and a first preset condition, thereby generating a second list of closed paths and determining the range of content values. This can automatically adjust the recognition steps from analyzing the belonging closed path to analyzing adjacent closed paths, which is beneficial to improving the fault tolerance capability of the recognized content value range.

[0137] Figure 4 This is an exemplary flowchart illustrating the determination of the scope of a title block according to some embodiments of this specification.

[0138] In some embodiments, process 400 can be executed by the image label information recognition system 100. For example... Figure 4 As shown, process 400 includes the following steps:

[0139] Step 410: Generate a list of graphic element information based on the text graphic elements in the drawing.

[0140] For more information on generating primitive information lists based on text primitives, please refer to [link / reference]. Figure 2 Related descriptions.

[0141] Step 420: Determine the aggregation range list based on the aggregation of text elements in the element information list.

[0142] Aggregation refers to the overall overlap between text elements. For example, the aggregation between two text elements can include overlap, partial overlap, containment, partial containment, and no overlap.

[0143] In some embodiments, the processor can determine the aggregation of text primitives based on the intersection-union ratio (CIRR). The CIRR is the ratio of the intersection to the union of the primitive ranges of multiple text primitives. For example, the processor can calculate the CIRR between the primitive ranges of any two text primitives and determine the aggregation by checking whether the CIRR falls within a threshold range. The threshold range can include multiple sub-threshold ranges, each corresponding to a different aggregation. When the CIRR of two text primitives falls within a certain sub-threshold range, the aggregation corresponding to that sub-threshold range is considered the aggregation of the primitive ranges of those two text primitives. In some embodiments, the correspondence between different sub-threshold ranges and different aggregation scenarios can be pre-defined.

[0144] An aggregation range list is a list that contains aggregation ranges. In some embodiments, an aggregation range may include the range of a single text primitive, the union of the ranges of multiple text primitives, etc. In some embodiments, the aggregation range list may be empty, in which case the aggregation range list does not contain the range of any text primitive.

[0145] In some embodiments, the processor may determine an aggregate range list based on the element extent and frame extent of the text primitives. Further details on how the aggregate range list is determined can be found in the following description.

[0146] Step 430: Determine the label range based on the aggregated range list.

[0147] In some embodiments, the processor can determine the tag range based on the frequency of occurrence of different tag fields in the aggregation range list. For example, the processor can determine the aggregation range corresponding to the tag field that appears most frequently as the tag range.

[0148] For example, the processor can calculate the intersection of other element ranges with the initial element range based on the four corner positions of the drawing (e.g., the lower left or lower right element range). If there is an intersection between the other element ranges and the initial element range, the processor takes the union of the other element ranges and the initial element range to obtain the first aggregate range. The processor then calculates the intersection of the first aggregate range with other element ranges. If there is an intersection, the processor takes the union of the first aggregate range with the other element ranges to obtain the second element range, and so on, until the last element range is merged into the nth aggregate range. The processor can then construct an aggregate range list based on the first to nth aggregate ranges, and determine the aggregate range containing the most title tag fields in the aggregate range list as the title tag range.

[0149] In some embodiments, for each text element in the element information list, the processor may further calculate a first intersection-union ratio (IUR) between the element extent of the text element and the frame extent of the frame; delete the text element from the element information list in response to the first IUR being greater than a third threshold; fill the element extent of the text element into the aggregation range list in response to the aggregation range list being empty; determine the union of each aggregation range in the aggregation range list with the element extent of the text element in response to the aggregation range list being non-empty, and calculate a second IUR between the union and the frame; fill the element extent of the text element into the aggregation range list in response to the second IUR being greater than a fourth threshold; replace the aggregation range with the union in response to the second IUR being less than or equal to the fourth threshold; and determine the tag range based on the number of occurrences of the tag field in at least one aggregation range in the aggregation range list.

[0150] The first intersection-to-union ratio (ICR) refers to the intersection-to-union ratio between the text primitive's extent and the bounding box's extent. The ICR can be represented numerically; the larger the value, the greater the overlap between the primitive's extent and the bounding box's extent.

[0151] In some embodiments, the processor may determine a first intersection-union ratio based on the size of the intersection and union between the primitive extent of the text primitive and the frame extent.

[0152] The third threshold is a threshold related to the first crossover ratio (CUP). For example, the third threshold can be any value between 0.1 and 0.2.

[0153] In some embodiments, the third threshold may be a system default value, an empirical value, a manually preset value, or any combination thereof, and can be set according to actual needs.

[0154] The second intersection-union ratio (IUU) is the intersection-union ratio between the union of the text primitive's primitive range and the aggregate range, and the bounding box range. The second IUU can be represented numerically; the larger the value, the greater the overlap between the union and the bounding box range.

[0155] In some embodiments, the processor may determine a second intersection-union ratio based on the union between the primitive extents and the aggregate extents of the text primitives, and the intersection and union sizes between the frame extents.

[0156] The fourth threshold is a threshold related to the second crossover ratio (CUP). For example, the fourth threshold can be any value between 0.1 and 0.4.

[0157] In some embodiments, the fourth threshold may be a system default value, an empirical value, a manually preset value, or any combination thereof, and can be set according to actual needs.

[0158] In some embodiments, the processor may determine the aggregate range in which the label field appears most frequently as the label range based on the number of times the label field appears in at least one aggregate range in the aggregate range list.

[0159] As an example only, for each text element in the element information list, the processor can calculate a first intersection-union ratio (IUR) between the element's extent and the frame's extent. If the first IUR is greater than a third threshold, the text element is removed from the element information list. If the first IUR is less than or equal to the third threshold, it is further determined whether the aggregation range list is empty. When the aggregation range list is empty (i.e., no aggregation range exists in the aggregation range list), the element's extent of the text element is added to the aggregation range list. When the aggregation range list is not empty (i.e., at least one aggregation range exists in the aggregation range list), the union of each aggregation range in the aggregation range list with the element's extent of the text element is determined, and a second IUR between the union and the frame is calculated. The relationship between the second IUR and a fourth threshold is determined. If the second IUR is greater than the fourth threshold, the element's extent of the text element is added to the aggregation range list. If the second IUR is less than or equal to the fourth threshold, the aggregation range is replaced with the union. Finally, the number of times the tag field appears in each aggregation range in the aggregation range list is calculated, and the aggregation range corresponding to the tag field with the most occurrences is determined as the tag range.

[0160] In some embodiments of this specification, at least one aggregation range is determined based on a first intersection-union ratio and a second intersection-union ratio, and then an aggregation range list is determined based on the aggregation range, thereby determining the tag range. This is beneficial for quickly eliminating text primitive ranges with excessive overlap, improving recognition efficiency, and reducing misjudgments.

[0161] In some embodiments of this specification, an aggregation range list is determined based on the aggregation of text elements in the element information list, thereby determining the tag range. This facilitates a more comprehensive identification of tag fields, a more accurate determination of the tag range, and simplifies the subsequent tag information extraction and parsing process.

[0162] It should be noted that the above description of process 400 is for illustrative purposes only and does not limit the scope of this specification. Those skilled in the art can make various modifications and changes to process 400 under the guidance of this specification. However, these modifications and changes remain within the scope of this specification.

[0163] The basic concepts have been described above. Obviously, for those skilled in the art, the detailed disclosure above is merely illustrative and does not constitute a limitation of this specification. Although not explicitly stated herein, those skilled in the art may make various modifications, improvements, and corrections to this specification. Such modifications, improvements, and corrections are suggested in this specification and therefore remain within the spirit and scope of the exemplary embodiments described herein.

[0164] Furthermore, this specification uses specific terms to describe embodiments thereof. For example, "an embodiment," "one embodiment," and / or "some embodiments" refer to a particular feature, structure, or characteristic associated with at least one embodiment of this specification. Therefore, it should be emphasized and noted that references to "an embodiment," "one embodiment," or "an alternative embodiment" in different locations throughout this specification do not necessarily refer to the same embodiment. Moreover, certain features, structures, or characteristics in one or more embodiments of this specification can be appropriately combined.

[0165] Furthermore, unless expressly stated in the claims, the order of processing elements and sequences, the use of numbers and letters, or other names described in this specification are not intended to limit the flow and sequence of the laminar flow shield. Although various examples have been discussed in the foregoing disclosure of some embodiments that are currently considered useful, it should be understood that such details are for illustrative purposes only, and the appended claims are not limited to the disclosed embodiments; rather, the claims are intended to cover all modifications and equivalent combinations that conform to the substance and scope of the embodiments described in this specification. For example, while the system components described above can be implemented using hardware devices, they can also be implemented solely using software solutions, such as installing the described system on existing servers or mobile devices.

[0166] Similarly, it should be noted that, in order to simplify the description disclosed herein and thus aid in the understanding of one or more embodiments of the invention, the foregoing description of embodiments in this specification may sometimes combine multiple features into a single embodiment, drawing, or description thereof. However, this method of disclosure does not imply that the subject matter of this specification requires more features than those mentioned in the claims. In fact, the embodiments contain fewer features than all the features of a single embodiment disclosed above.

[0167] In some embodiments, numbers describing the quantity of components and attributes are used. It should be understood that such numbers used in the description of embodiments are modified in some examples with the terms "approximately," "approximately," or "generally." Unless otherwise stated, "approximately," "approximately," or "generally" indicates that the numbers are allowed to vary by ±20%. Accordingly, in some embodiments, the numerical parameters used in the specification and claims are approximate values, which may be changed depending on the characteristics required by individual embodiments. In some embodiments, numerical parameters should take into account specified significant digits and employ a general method of digit reservation. Although the numerical ranges and parameters used to confirm their breadth of range in some embodiments of this specification are approximate values, in specific embodiments, such values ​​are set as precisely as feasible.

[0168] For each patent, patent application, patent application publication, and other material, such as articles, books, specifications, publications, and documents, referenced in this specification, the entire contents of which are incorporated herein by reference. This excludes historical application documents that are inconsistent with or conflict with the content of this specification, as well as documents that limit the broadest scope of the claims in this specification (currently or subsequently appended to this specification). It should be noted that in the event of any inconsistency or conflict between the descriptions, definitions, and / or terminology used in the supplementary materials to this specification and the content of this specification, the descriptions, definitions, and / or terminology used in this specification shall prevail.

[0169] Finally, it should be understood that the embodiments described in this specification are merely illustrative of the principles of the embodiments described herein. Other variations may also fall within the scope of this specification. Therefore, alternative configurations of the embodiments described herein are intended to be illustrative rather than limiting, and should be considered consistent with the teachings of this specification. Accordingly, the embodiments described herein are not limited to those explicitly introduced and described herein.

Claims

1. A method for identifying image tag information, characterized in that, The method is executed by a processor and includes: Obtain the title block area of ​​the drawing; Based on the drawing frame and the label range of the drawing, a first closed path list is determined, which includes all first closed paths that do not contain other closed paths. Based on the text element range in the text element list and the first closed path in the first closed path list, determine the closed path to which the tag field belongs. Based on the first intersection ratio between the closed path and the primitive range, and the offset distance between the center point of the text primitive range and the center point of the closed path, the content value range of the tag field is determined; Based on the content value range, obtain the text content value within the content value range; The determination of the content value range of the tag field, based on the first intersection ratio between the closed path and the primitive range, and the offset distance between the center point of the text primitive's range and the center point of the closed path, includes: Determine whether the first intersection ratio and the offset distance satisfy a first preset condition; and In response to the first intersection ratio and the offset distance satisfying a first preset condition, the range of the closed path to which the content value is to be used as the range of the content value; In response to the first intersection ratio and / or the offset distance not meeting the first preset condition: Determine the adjacent closed paths that share a vertex with the closed path to which it belongs; The adjacent closed paths that meet the filtering criteria are identified as second closed paths, and a list of second closed paths is generated; and Based on the second list of closed paths, the range of content values ​​is determined, including: Determine the number of paths in the second closed path list; In response to the fact that the number of paths in the second closed path list is 0, the closed path is determined to be within the range of content values; In response to the fact that the number of paths in the second closed path list is greater than 0, a second intersection ratio between the second closed path and the label range is determined; In response to the fact that the number of second closed paths in the second closed path list that satisfy a second preset condition regarding the second intersection ratio with the label range is less than a preset number, the second closed paths whose second intersection ratio satisfies the second preset condition are used as the content value range; and In response to the fact that the number of second closed paths in the second closed path list that have a second intersection ratio with the tag range that satisfies the second preset condition is not less than a preset number, the target distance between the center point of each second closed path and the center point of the corresponding closed path in a preset direction is calculated, and the second closed path whose target distance with the center point of the corresponding closed path in the preset direction satisfies the third preset condition is taken as the content value range.

2. The method according to claim 1, characterized in that, The first preset condition includes that the first intersection ratio is less than or equal to a first threshold, and the ratio of the offset distance to the target direction length of the closed path is greater than a second threshold.

3. The method according to claim 1, characterized in that, The determination of the first closed path list based on the drawing frame and the title block range includes: Determine the straight line segments that intersect the frame and the label area; An undirected graph is constructed based on the line segments and the intersection of the frame and the label area; Traverse the undirected graph to determine multiple third closed paths; and The first list of closed paths is determined based on the inclusion relationships among the plurality of third closed paths.

4. The method according to claim 1, characterized in that, The list of text elements is obtained through the following steps: Based on the text elements in the drawing, generate an element information list; and Based on the graphic element information list and the commonly used text information table, the text graphic element list is generated.

5. The method according to claim 1, characterized in that, The title page range of the drawing is obtained through the following steps: Based on the text elements in the drawing, generate an element information list; Based on the aggregation of text elements in the element information list, an aggregation range list is determined, wherein the aggregation range list includes the element range of at least one of the text elements; and The scope of the image tag is determined based on the aggregated scope list.

6. The method according to claim 5, characterized in that, Based on the aggregated range list, the scope of the image tag is determined to include: For each text element in the element information list, calculate the first intersection-union ratio between the element range of the text element and the frame range of the frame; In response to the first intersection-union ratio being greater than a third threshold, the text primitive is removed from the primitive information list; In response to the first intersection-union ratio being less than or equal to the third threshold, In response to the fact that the aggregation range list is empty, the element range of the text element is filled into the aggregation range list; In response to the fact that the list of aggregate ranges is not empty, the union of each aggregate range in the list of aggregate ranges with the range of the text primitive is determined, and the second intersection-union ratio of the union with the frame is calculated; In response to the second intersection-union ratio being greater than the fourth threshold, the primitive range of the text primitive is filled into the aggregation range list; In response to the second intersection-union ratio being less than or equal to the fourth threshold, the aggregation range is replaced with the union set; and The tag range is determined based on the number of times the tag field appears in at least one aggregation range in the aggregation range list.

7. A label information recognition system, characterized in that, The system is used to implement the image tag information recognition method as described in any one of claims 1-6, and the system includes: The acquisition module is configured to acquire the title block range of the drawing; The first determining module is configured to determine a first closed path list based on the drawing frame and the label range of the drawing, wherein the first closed path list includes all first closed paths that do not contain other closed paths; The second determining module is configured to determine the closed path to which the tag field belongs based on the element range of the text elements in the text element list and the first closed path in the first closed path list. The third determining module is configured to determine the content value range of the tag field based on a first intersection ratio between the closed path and the primitive range, and the offset distance between the center point of the text primitive's range and the center point of the closed path; and The extraction module is configured to obtain text content values ​​within the content value range based on the content value range.

Images