Information processing systems, information processing methods, and programs
The system efficiently extracts objects from drawings by recognizing character positions and orientations to define reference areas, enabling accurate object division and extraction for database creation and inspection.
Patent Information
- Authority / Receiving Office
- JP · JP
- Patent Type
- Applications
- Current Assignee / Owner
- OHBAYASHI GUMI LTD
- Filing Date
- 2024-11-06
- Publication Date
- 2026-05-19
AI Technical Summary
Existing information processing systems fail to efficiently extract objects from drawings using characters within the drawings.
An information processing system that includes a position and angle information acquisition unit to recognize character positions and orientations, a reference area extraction unit to define areas based on character placement, and a target area extraction unit to identify the object area corresponding to the reference area.
Enables efficient extraction of objects from drawings by utilizing characters, allowing for accurate division and extraction of individual objects, facilitating database creation, inspection, and similar part searching.
Smart Images
Figure 2026082103000001_ABST
Abstract
Description
Technical Field
[0006] , , ,
[0005] , , ,
[0001] The present disclosure relates to an information processing system, an information processing method, and a program.
Background Art
[0002] Patent Document 1 discloses an information processing apparatus that divides line segment data of parts drawn on a screen into parts by extracting the line segment data included in a rectangular area using a rectangular area covering the line segment.
Prior Art Documents
Patent Documents
[0003]
Patent Document 1
Summary of the Invention
Problems to be Solved by the Invention
[0004] However, Patent Document 1 does not disclose a technique for extracting an object displayed in a drawing using characters included in the drawing.
[0005] Therefore, in one aspect, an object of the present invention is to provide an information processing system or the like that can efficiently extract an object displayed in a drawing using characters included in the drawing.
Means for Solving the Problems
[0006] In one aspect, a position and angle information acquisition unit that acquires the position and direction of characters indicating the dimensions of an object displayed in a drawing; a reference area extraction unit that extracts a predetermined reference area including the area where the characters are arranged based on the position and the direction acquired by the position and angle information acquisition unit; an object area extraction unit that extracts an object area that is an area corresponding to the object based on the reference area extracted by the reference area extraction unit; An information processing system is provided that includes the following features. [Effects of the Invention]
[0007] In one respect, according to the present invention, it is possible to efficiently extract objects shown in a drawing by utilizing the characters contained in the drawing. [Brief explanation of the drawing]
[0008] [Figure 1] This is a diagram showing the configuration of the information processing system in this embodiment. [Figure 2] This is an example diagram showing a drawing in which the target is displayed. [Figure 2A] This figure illustrates a reference region extracted by the reference region extraction unit. [Figure 2B] This is a partially enlarged view of Figure 2A. [Figure 3] This is a flowchart showing the process of dividing a drawing. [Figure 4] This diagram shows the arrangement of the target area on the drawing. [Modes for carrying out the invention]
[0009] Figure 1 shows the configuration of the information processing system in this embodiment.
[0010] As shown in Figure 1, the information processing system 10 of this embodiment includes a position and angle information acquisition unit 11 that acquires the position and orientation of characters indicating the dimensions of an object displayed on a drawing, a reference area extraction unit 12 that extracts a predetermined reference area including the area where the characters are placed based on the position and orientation acquired by the position and angle information acquisition unit 11, and a target area extraction unit 13 that extracts a target area 100 which is the area corresponding to the object based on the reference area extracted by the reference area extraction unit 12.
[0011] The information processing system 10 can be configured using one or more computers on which a predetermined program is implemented.
[0012] Next, the operation of the information processing system 10 will be described.
[0013] Figure 2 is an example of a drawing in which an object is displayed. In the object area 100 shown in Figure 2, for example, a front view, a side view, etc., showing the shape of the object component are displayed, and dimensions are indicated around it. In Figure 2, the drawing showing the shape of the component in the object area 100 is omitted. Also, the characters indicating dimensions on the drawing are marked with circles.
[0014] The position and angle information acquisition unit 11 recognizes characters (strings) such as numbers indicating dimensions on the drawing and acquires the position and direction of the characters. For example, the position and angle information acquisition unit 11 can extract numbers indicating dimensions from character information embedded in the drawing or from an OCR of the drawing image, and recognize their coordinates and direction.
[0015] The position and angle information acquisition unit 11 acquires, for example, the position of the string 101a located above the target area 100 in the diagram shown in Figure 2. The position and angle information acquisition unit 11 also acquires the X direction, which is the horizontal orientation of the string 101a, as the direction of the string 101a. Similarly, the position and angle information acquisition unit 11 acquires the position of the string 101b located below the target area 100. The position and angle information acquisition unit 11 also acquires the X direction, which is the horizontal orientation of the string 101b, as the direction of the string 101b.
[0016] Furthermore, the position and angle information acquisition unit 11 acquires the position of the string 102a located to the left of the target area 100 in the diagram shown in Figure 2. The position and angle information acquisition unit 11 also acquires the Y direction, which is the horizontal orientation of the string 102a, as the direction of the string 102a. Similarly, the position and angle information acquisition unit 11 acquires the position of the string 102b located to the right of the target area 100. The position and angle information acquisition unit 11 also acquires the Y direction, which is the horizontal orientation of the string 102b, as the direction of the string 102b.
[0017] Based on the positions and orientations of the character strings 101a, 101b, 102a, and 102b acquired by the position and angle information acquisition unit 11, the reference area extraction unit 12 extracts a predetermined reference area that includes the area where the characters are arranged.
[0018] FIG. 2A is a diagram illustrating a reference area extracted by the reference area extraction unit, and FIG. 2B is a partially enlarged view of FIG. 2A.
[0019] In the examples of FIGS. 2A and 2B, based on the range where a group of character strings 101A in the X direction including the character string 101a above the target area 100 exists, a reference area 110Y is extracted.
[0020] In this example, the reference area 110Y is extracted as a strip-shaped area having a width in the X direction that includes the range where the group of character strings 101A exists and extending in the Y direction. When extracting the reference area 110Y, in addition to the range where the group of character strings 101A exists, dimension lines located in the vicinity of the group of character strings 101A may be referred to. For example, in the example of FIG. 2B, the range of the reference area 110Y may be set by referring to the positions of the dimension lines 105L and 105R extending in the Y direction and the dimension line 103 extending in the X direction between the dimension lines 105L and 105R. In this case, an area including the dimension lines 105L, 105R, and 103 that extends outside the range of the group of character strings 101A in the X direction can be extracted as the reference area 110Y.
[0021] Note that in FIG. 2B, an example of extracting the reference area 110Y based on the range where a group of character strings 101A above the target area 100 exists is shown, but the reference area 110Y can also be extracted based on the range where a group of character strings 101B below the target area 100 exists. Also, when the range of the extracted reference area 110Y is different between the case based on the range where the group of character strings 101A exists and the case based on the range where the group of character strings 101B exists, the entire area including at least one of them can be extracted as the final reference area 110Y. Thereby, the possibility that a part of the target area 100 is excluded from the reference area 110Y can be eliminated.
[0022] Similarly, the reference region 110X is extracted based on the range in which a group of strings 102A in the Y direction, including string 102a located to the left of the target region 100, and a group of strings 102B in the Y direction, including string 102b located to the right of the target region 100, exist.
[0023] The reference region 110X is extracted as a band-shaped region that has a width in the Y direction that includes the range where the group of strings 102A and 102B exist, and extends in the X direction. Similar to the extraction of the reference region 110Y, when extracting the reference region 110X, in addition to the range where the group of strings 102A and 102B exist, dimension lines located near the group of strings 102A and 102B may also be referenced. In this case, the region including dimension lines that extend outward in the Y direction beyond the range of the group of strings 101A and 102B can be extracted as the reference region 110X.
[0024] The target region extraction unit 13 extracts the target region 100, which is the region corresponding to the target, based on the reference regions 110X and 110Y extracted by the reference region extraction unit 12.
[0025] For example, the target area extraction unit 13 may extract the area where the reference area 110X and the reference area 110Y overlap as the target area 100, or it may extract a part of the area where the reference area 110X and the reference area 110Y overlap as the target area 100. For example, in the latter case, the area where the reference area 110X and the reference area 110Y overlap and where it is determined that there are lines indicating the shape of the member based on the image data of the drawing may be extracted as the target area 100. Alternatively, the target area 100 may be an expanded area that includes part or all of the group of strings 101A, 101B and the group of strings 102A, 102B.
[0026] Next, the process of dividing a drawing using the information processing system 10 will be described. This process shows an example of dividing a drawing in which multiple target areas 100A to 100E are drawn into each of the target areas 100A to 100E.
[0027] Figure 3 is a flowchart showing the process of dividing a drawing, and Figure 4 is a diagram showing the arrangement of target areas on the drawing.
[0028] In step S102 of Figure 3, the position and angle information acquisition unit 11 extracts one of the group of strings (corresponding to the group of strings 101A, 101B and the group of strings 102A, 102B in Figure 2A) and acquires the position and angle of the character (corresponding to string 101a, etc.) included in the group of strings.
[0029] In step S104, the reference area extraction unit 12 acquires line segments 103A (dimension line, corresponding to dimension line 103 in Figure 2B, etc.) and 104A shown in Figure 4, which are located near the group of characters extracted in step S102 and match the angle of the characters acquired in step S102. Here, the reference area extraction unit 12 acquires all line segments that satisfy this condition.
[0030] In step S106, the reference region extraction unit 12 obtains the positions of both ends of the line segments (for example, line segments 103A and 104A) acquired in step S104. Furthermore, the reference region extraction unit 12 extends the line segments by Δd in both directions corresponding to the angle (see Figure 2B). As a result, for example, the width of the strip of the reference region 111Y shown in Figure 4 (corresponding to the reference region 110Y in Figures 2A and 2B), which has a strip shape, is set by the ends of the extended line segments.
[0031] In step S108, the reference region extraction unit 12 extracts a band-shaped region with a set width extended in the Y direction as the reference region 111Y (Figure 4).
[0032] In step S110, the reference area extraction unit 12 determines whether all strings have been extracted. If the determination is positive, the process proceeds to step S112; otherwise, the process proceeds to step S102.
[0033] Through the above process, it is possible to extract reference regions such as 111X, 112X, 113X, 111Y, and 112Y, as shown in Figure 4.
[0034] In step S112, the target region extraction unit 13 extracts one candidate target region. Here, the target region extraction unit 13 extracts a region where reference regions such as 111X, 112X, 113X, 111Y, and 112Y overlap.
[0035] For example, the region containing the target region 100A shown in Figure 4 is extracted as the region where the mutually intersecting reference regions 111X and 111Y overlap. Similarly, the region containing the target region 100B is extracted as the region where the mutually intersecting reference regions 111X and 112Y overlap, and the region containing the target region 100E is extracted as the region where the mutually intersecting reference regions 112X and 112Y overlap.
[0036] In step 114, the target region extraction unit 13 uses a predetermined algorithm to determine whether or not the shape of a component is drawn in the candidate target region extracted in step S112. If the determination is affirmative, the target region extraction unit 13 proceeds to step S116; otherwise, it proceeds to step S112.
[0037] In step 116, the target region extraction unit 13 obtains bounding boxes corresponding to the extracted candidate reference regions. That is, the target region extraction unit 13 extracts the target region in which the component is drawn.
[0038] In step 118, the target area extraction unit 13 determines whether or not all candidate target areas on the drawing have been extracted. If the determination is negative, the process ends; otherwise, the process proceeds to step S112.
[0039] Through the processing in steps S112 to S118, for example, all target regions 100A, 100B, 100C, 100D, 100E, and 100F shown in Figure 4, i.e., all regions where the members are drawn, are extracted independently. This allows the drawing to be divided into target regions 100A to 100F.
[0040] Furthermore, when the same object (component, etc.) is assigned to each target area, such as a front view, side view, top view, and rear view, the target areas are arranged in the X or Y direction, as shown in Figure 4. For example, in Figure 4, target areas 100A, 100B, 100C, and 100D are arranged in the X direction, while target areas 100B, 100E, and 100F are arranged in the Y direction. Therefore, the arrangement of the target areas can be used when extracting target areas. For example, if target area 100B has already been extracted, candidate target areas can be efficiently extracted by preferentially searching for areas shifted in the X or Y direction relative to target area 100B.
[0041] As described above, according to this embodiment, objects displayed in a drawing can be efficiently extracted by using the characters indicating dimensions written on the object. This makes it possible to divide and extract individual objects from a drawing containing various information, and the extraction results can be used, for example, for creating a database of drawing information (part shape, item), inspection, and searching for similar parts. Furthermore, when extracting objects such as parts from a drawing, even if there is insufficient space between objects, the object area in which each object is drawn can be accurately extracted.
[0042] In the above embodiment, a component is used as an example of the target, but the target is arbitrary. For example, the target could be a building, a structure, etc. Furthermore, the size of the drawing paper, the scale indicated on the drawing, and the attributes of the target may also be used to extract the target area.
[0043] Although the embodiments have been described in detail above, the invention is not limited to any particular embodiment, and various modifications and changes are possible within the scope of the claims. Furthermore, it is possible to combine all or more of the components of the embodiments described above. [Explanation of symbols]
[0044] 10. Information Processing Systems 11 Position angle information acquisition unit 12 Reference area extraction part 13 Target area extraction unit Target areas: 100, 100A, 100B, 100C, 100D, 100E, 100F 101a, 101b string 102a, 102b string 110X reference area 110Y reference area
Claims
1. A position and angle information acquisition unit that acquires the position and orientation of the characters indicating the dimensions of an object displayed on a drawing, A reference area extraction unit extracts a predetermined reference area that includes the area where the characters are placed, based on the position and direction acquired by the position and angle information acquisition unit. A target region extraction unit extracts a target region that corresponds to the target based on the reference region extracted by the reference region extraction unit, An information processing system equipped with the following features.
2. The information processing system according to claim 1, wherein the reference region extraction unit extracts a band-shaped region as the reference region, which includes the position of the character acquired by the position and angle information acquisition unit and extends in a direction perpendicular to the direction of the character acquired by the position and angle information acquisition unit.
3. A position and angle information acquisition step to obtain the position and orientation of the text indicating the dimensions of an object displayed on a drawing, A reference area extraction step extracts a predetermined reference area that includes the area where the characters are placed, based on the position and direction obtained in the position and angle information acquisition step, A target region extraction step, which extracts a target region that corresponds to the target, based on the reference region extracted in the reference region extraction step, An information processing method comprising:
4. A position and angle information acquisition step to obtain the position and orientation of the text indicating the dimensions of an object displayed on a drawing, A reference area extraction step extracts a predetermined reference area that includes the area where the characters are placed, based on the position and direction obtained in the position and angle information acquisition step, A target region extraction step, which extracts a target region that corresponds to the target, based on the reference region extracted in the reference region extraction step, A program that causes a computer to execute something.