Apparatus, method, and program for managing drawing data including raster data.

Abstract

To provide a device, method and program for recognizing a value of dimension within a drawing represented by drawing data for managing the drawing data including raster data.SOLUTION: A device acquires drawing data of a raster type (S201), vectorizes the drawing data to be converted into vector data (S202), generates dimension line data associated with first and second nodes included in the vector data (S203), specifies, from the vector data, one or more coupled edges adjacent to a dimension line represented by the dimension line data (S204), applies character recognition to a corresponding region within the drawing data corresponding to a region including the specified one or more coupled edges (S205), and stores a recognized character as a dimension in association with the dimension line data (S206).SELECTED DRAWING: Figure 2

Description

Technical Field

[0001] The present invention relates to an apparatus, a method, and a program therefor for managing drawing data including raster data.

Background Art

[0002] In the manufacturing industry, as there are about 30,000 parts in one automobile, the required parts are diverse. Generally, manufacturers do not process all these various parts by themselves, but utilize outsourcing production to external processors.

[0003] In outsourcing production, when a manufacturer entrusts production, the manufacturer provides a part drawing representing the part to be entrusted to the processor in an electronic medium or a paper medium printed from the electronic medium for estimation, ordering, etc. The electronic medium is mainly managed as data in PDF format. Also, a processor that receives a part drawing on paper scans it and mainly manages it as data in PDF format.

Summary of the Invention

Problems to be Solved by the Invention

[0004] PDF format drawing data includes those with layers and those without layers. In the former case, some of the lines, characters, and symbols in the drawing represented by the drawing data may be distinguishable by being represented by data in layers according to their respective types.

[0005] However, the PDF format drawing data managed as described above is often obtained, for example, by printing and then scanning after adding handwritten characters or symbols, and often does not have layers. Therefore, information regarding the types of lines, characters, and symbols in the drawing is often not retained.

[0006] In manufacturing, drawing data representing parts includes not only shape and dimensions, but also data related to part name, drawing number, serial number, material, surface treatment, etc. For example, if it is possible to recognize what a certain number represents, various advantages can be expected in software used to manage drawing data, such as improved searchability and easier extraction of similar drawings. The advantages are particularly significant if dimensional values ​​can be distinguished from other numerical values.

[0007] Although the explanation used PDF image data as an example, the present invention is broadly applicable to drawing data representing parts that includes raster data, and is particularly suitable for data without layers. Examples include data formats such as JPG and TIFF. Raster data is data that includes multiple pixels. JPG or TIFF image data generally contains only raster data, but PDF image data may contain vector data or text data in addition to raster data.

[0008] The present invention has been made in view of these points, and its purpose is to enable recognition of dimensional values ​​in a drawing represented by drawing data in an apparatus, method, and program for managing drawing data including raster data. [Means for solving the problem]

[0009] To achieve this objective, a first aspect of the present invention is a method for managing drawing data including raster data, comprising the steps of: vectorizing the drawing data to generate vector data; generating dimension line data associated with first and second nodes included in the vector data; performing character recognition on corresponding areas in the drawing data that correspond to adjacent areas adjacent to dimension lines represented by the dimension line data; and storing the characters obtained by the character recognition as dimension values ​​associated with the dimension line data.

[0010] Furthermore, a second aspect of the present invention is the method of the first aspect, wherein the step of generating dimension line data includes the steps of detecting a plurality of tip vertex symbols from the drawing data, determining at least one pair of tip vertex symbols among the detected plurality of tip vertex symbols, and identifying first and second nodes in the vector data corresponding to the first and second tip vertices of the pair of tip vertex symbols.

[0011] Furthermore, a third aspect of the present invention is the method of the second aspect, wherein the tip vertex symbol is an arrow.

[0012] Furthermore, a fourth aspect of the present invention is a method according to any of the first to third aspects, further comprising the step of determining the orientation of the adjacent region or the corresponding region.

[0013] Furthermore, a fifth aspect of the present invention is a method according to any of the first to fourth aspects, wherein the proximity region is a region that includes one or more connected edges adjacent to the dimension line represented by the dimension line data.

[0014] Furthermore, a sixth aspect of the present invention is the method of the fifth aspect, wherein the proximity region is a region including one or more connected edges located within a predetermined range in a direction perpendicular to the edge connecting the first and second nodes or a line segment extending beyond one of the nodes of the edge.

[0015] Furthermore, a seventh aspect of the present invention is a method according to any of the first to sixth aspects, further comprising the step of calculating the number of pixels per millimeter in the drawing represented by the drawing data using the dimensional values.

[0016] Furthermore, an eighth aspect of the present invention is a method according to any of the first to seventh aspects, further comprising the step of performing a thinning process on the drawing data before vectorization.

[0017] Furthermore, a ninth aspect of the present invention is a program for causing a computer to perform a method for managing drawing data including raster data, the method comprising the steps of: vectorizing the drawing data to generate vector data; generating dimension line data associated with first and second nodes included in the vector data; performing character recognition on corresponding areas in the drawing data that correspond to adjacent areas adjacent to dimension lines represented by the dimension line data; and storing the characters obtained by the character recognition as dimension values ​​associated with the dimension line data.

[0018] Furthermore, a tenth aspect of the present invention is a device for managing drawing data including raster data, which vectorizes the drawing data to generate vector data and dimension line data associated with first and second nodes included in the vector data, performs character recognition on corresponding areas in the drawing data that correspond to adjacent areas adjacent to dimension lines represented by the dimension line data, and stores the characters obtained by the character recognition as dimension values ​​associated with the dimension line data. [Effects of the Invention]

[0019] According to one aspect of the present invention, drawing data including raster data is vectorized to generate dimension line data associated with first and second nodes included in the vector data, character recognition is performed on corresponding areas in the drawing data that correspond to adjacent areas adjacent to the dimension lines represented by the dimension line data, and the obtained characters are associated with the dimension line data and stored as dimension values, thereby making it possible to recognize the dimension values ​​in the drawing represented by the drawing data. [Brief explanation of the drawing]

[0020] [Figure 1] This figure shows a device for managing drawing data according to a first embodiment of the present invention. [Figure 2] This figure shows the flow of a method for managing drawing data according to the first embodiment of the present invention. [Figure 3] FIG. 1 is a partially enlarged view of an example of a drawing displayed based on drawing data according to the first embodiment of the present invention. [Figure 4] FIG. 2 is a diagram showing an example of dimension line data according to an embodiment of the present invention. [Figure 5] FIG. 3 is a diagram in which a plurality of connected edges specified as being close to the dimension lines shown in the drawing of FIG. 3 are drawn on the drawing. [Figure 6] FIG. 4 is a diagram showing an example in which a dimension value is added to dimension line data according to an embodiment of the present invention. [Figure 7] FIG. 5 is a diagram showing the flow of a method for detecting dimension lines according to an embodiment of the present invention. [Figure 8] FIG. 6 is a diagram in which a plurality of arrows included in the drawing of FIG. 3 are drawn emphasized on the drawing. [Figure 9] FIG. 7 is a diagram showing an example of data describing the arrows included in the drawing of FIG. 3.

Mode for Carrying Out the Invention

[0021] Hereinafter, embodiments of the present invention will be described in detail with reference to the drawings.

[0022] (First Embodiment) FIG. 1 shows an apparatus for managing drawing data according to the first embodiment of the present invention. The apparatus 100 communicates with a manufacturer terminal 110 used by a manufacturer via an IP network such as the Internet, enabling the management of drawing data. Further, the apparatus 100 can similarly communicate with a processor terminal 120 used by a processor and provide a drawing management service to the processor. The user of the drawing management service is not particularly limited, but in this embodiment, an example in which a manufacturer uses the drawing management service will be described.

[0023] The device 100 comprises a communication unit 101 such as a communication interface, a processing unit 102 such as a processor or CPU, and a storage unit 103 including a storage device or storage medium such as memory or a hard disk, and can be configured by executing a program for each processing. The device 100 may include one or more devices, computers or servers. The program may also include one or more programs and can be recorded on a computer-readable storage medium to become a non-transient program product. The program can be stored in a storage device or storage medium such as a database 104 accessible from the storage unit 103 or the device 100 via an IP network, and can be executed by the processing unit 102. The data described below as being stored in the storage unit 103 may also be stored in the database 104, and vice versa.

[0024] First, the device 100 acquires drawing data in raster format (S201). For example, this drawing data is stored in the storage unit 103 by being uploaded and registered by the manufacturer's purchasing manager from a web page provided by the device 100, and is acquired by reading it. Figure 3 is a partial enlargement view of an example of a drawing displayed using the drawing data.

[0025] Next, the device 100 vectorizes the drawing data and converts it into vector data (S202). In this specification, "vectorization" means converting the data into data that represents multiple points (hereinafter also referred to as "nodes") and one or more line segments (hereinafter also referred to as "edges") that connect at least two of the multiple points. The obtained vector data may be in a separate file from the drawing data, or a layer may be added to the file containing the drawing data, and the data may be in a different layer from the drawing data.

[0026] In the vector data in question, each edge can be represented as follows, for example: [ [x1, y1], [x2, y2] ] Here, the lower left corner of the drawing, represented by the drawing data, is taken as the origin, with the x-axis running horizontally and the y-axis running vertically, and each node of the edge is represented in the format [x,y]. The line segments drawn in Figure 3 may not necessarily be single edges as vector data, but may be converted into multiple edges.

[0027] The drawing data may be thinned before vectorization. For example, by binarizing or multi-leveling each pixel of the raster data included in the drawing data as needed, and narrowing the width of continuous areas that have the same value, the precision of nodes and edges can be improved.

[0028] Next, the device 100 generates dimension line data associated with the first and second nodes included in the vector data (S203). Figure 4 shows an example of dimension line data according to one embodiment of the present invention. The dimension line data is in JSON format representing a dimension line with a diameter of 24 mm as shown in Figure 3. "Dimension" indicates that it is dimension line data, and the value of the key "dimension_chains" describes information about the dimension line. The value of the key "projection_lines" describes two dimension extension lines associated with the dimension line. The value of the key "dimension_chains" includes the coordinates of the first node "p1", which is one of the apex ends of the detected dimension line, and the coordinates of the second node "p2", which is the other apex end. The key "arrow_heads" represents the direction of the first and second arrows on the dimension line. Details of dimension line detection will be described later.

[0029] Subsequently, the device 100 identifies one or more connected edges adjacent to the dimension line represented by the dimension line data from the vector data (S204). More specifically, it is preferable that edges that intersect the edges connecting the first and second nodes of the dimension line data, or nodes or edges that are too close to the edges connecting the first and second nodes of the dimension line data (for example, within 3 pixels), are not included in the one or more connected edges adjacent to the dimension line.

[0030] The device 100 then performs character recognition on the corresponding area in the drawing data corresponding to the adjacent area containing the identified connected edges (S205), and stores the recognized characters as dimensions associated with dimension line data (S206). When performing character recognition, the orientation of the characters can be determined to improve recognition accuracy if the orientation of the characters is known. For example, the orientation of the edge connecting the first and second nodes can be determined as the orientation of the adjacent area or the corresponding area. Alternatively, the orientation of the longer side of the adjacent area or the corresponding area can be determined as the orientation of the adjacent area or the corresponding area. Details of identifying one or more connected edges will be described later. Here, "connected" means that the edges have a common node.

[0031] Since connected edges adjacent to dimension lines are highly likely to represent dimensions, the dimension values ​​contained in the drawing data can be recognized in this way. Here, we identified one or more connected edges and focused on the corresponding area in the drawing data that corresponds to the area containing those one or more connected edges. However, even if we focus on the corresponding area in the drawing data that corresponds to the adjacent area adjacent to the dimension line represented by the dimension line data, the characters within that corresponding area are not unlikely to represent dimension values, and the dimension values ​​contained in the drawing data can be recognized.

[0032] While not limited to this, if the device 100 can store the dimensional values ​​included in the drawing data, it can, for example, enable users of the drawing management service provided by the device 100 to search the drawing data by specifying at least one of the maximum dimension, minimum dimension, and dimensional range included in the part.

[0033] Figure 5 shows multiple connected edges identified as being adjacent to the detected dimension line drawn on the drawing, and region 500 is identified as the region containing these edges. Figure 6 shows the coordinates of the four vertices of the region containing the identified one or more connected edges, the characters resulting from character recognition for that region, and the numerical values ​​represented by those characters added to the value of the key "dimensions" in the value of the key "dimension_chains". The key "pixel_per_millimeter" describes the result of calculating how many pixels correspond to 1 mm in this drawing data from the number of pixels of the dimension line and the dimension value of the dimension line. For example, for each of the generated dimension line data, the value of the number of pixels per 1 mm can be calculated from the dimension value associated with that dimension line data, and the median value of the multiple calculation result columns obtained can be used as the value of the key "pixel_per_millimeter". Using this value, it is possible to calculate the dimensions of any part in the component.

[0034] Details of dimension line detection Figure 7 shows the flow of a dimension line detection method according to one embodiment of the present invention. First, the device 100 detects multiple arrows from the drawing data by pattern recognition using image processing (S701). Figure 8 shows the detected multiple arrows highlighted on the drawing. Figure 9 shows an example of data describing the detected arrows. This data is written in JSON format, where the value of the key "head" represents the vertex coordinates of the arrow, and the value of the key "dir" represents the direction of the arrow. This arrow corresponds to the arrow above the dimension line with a diameter of 24 mm.

[0035] The device 100 then determines at least one pair of arrows from the detected plurality of arrows (S702) and identifies the first and second nodes in the vector data corresponding to the first and second tip vertices of the first and second arrows of the pair of arrows (S703). Once the first and second nodes corresponding to the tip vertices are identified, the device 100 can generate dimension line data associated with the first and second nodes.

[0036] By defining the same coordinate system for the drawing data as for the vector data, the first and second nodes corresponding to the coordinates of the first and second tip vertices can be identified. In this case, there are not necessarily nodes at coordinates that perfectly match the coordinates of the first and second tip vertices of the determined pair of arrows, but for example, nodes that are close to or nearest to the first and second tip vertices can be identified as the first and second nodes, respectively. Furthermore, the coordinate system defined for the vector data and the coordinate system defined for the drawing data do not need to be perfectly identical; it is sufficient that the coordinate points can be converted between them.

[0037] The vertices at the ends of dimension lines may be represented by symbols other than arrows, such as black dots. In this case, the substitute symbol is detected by image processing. The description of the key "arrow_heads" shown in Figure 4 is modified as appropriate to match the substitute symbol. The descriptions using arrow examples in this specification also apply similarly to vertex symbols in general, including arrows.

[0038] Specific details of one or more connected edges Regarding the 24mm diameter dimension line mentioned above, if one or more connected edges located within a predetermined range perpendicular to the edge connecting the first and second nodes are detected, there is a high probability that they correspond to the dimension value.

[0039] When the length of the line segment between nodes is short, such as in a dimension line with a length of 8 mm, the arrow may point inward. Although the example in Figure 8 is different, in such cases the dimension value may not be drawn perpendicular to the edge connecting the nodes, but rather close to the line segment that extends beyond one node of that edge. In such cases, it is necessary to detect one or more connected edges located within a predetermined range perpendicular to the extended line segment.

[0040] By excluding one or more edges adjacent to a dimension line represented by dimension line data that are not adjacent if their vertical distance from an edge connecting nodes in the dimension line data, or from a line segment extending that edge beyond one node, is greater than or equal to a predetermined value from the median of multiple estimated distance results obtained for those edges, the accuracy of identifying adjacent regions and their corresponding regions can be improved.

[0041] Furthermore, in the embodiments described above, unless the word "only" is used, such as "based only on," "depending only on," or "in the case of only," it is assumed in this specification that additional information may also be considered. Also, as an example, the statement "if a, then b" does not necessarily mean "always b in the case of a" or "b immediately after a," unless explicitly stated otherwise. In addition, the statement "each a constituting A" does not necessarily mean that A is composed of multiple components, but includes the possibility that the component is singular.

[0042] Furthermore, for the sake of clarity, even if there are aspects of operation in some method, program, terminal, device, server, or system (hereinafter referred to as "method, etc.") that differ from the operation described herein, each aspect of the present invention is intended to cover the same operation as any of the operations described herein, and the existence of operation different from the operation described herein does not mean that such method, etc. is outside the scope of each aspect of the present invention.

[0043] Furthermore, the "start" and "end" shown in Figure 2 are merely examples, and do not necessarily mean that the method according to this embodiment always starts at S201 and always ends at S206. [Explanation of symbols]

[0044] 100 devices 101 Communications Department 102 Processing Unit 103 Storage section 104 Databases 110 Manufacturer terminal 120 Processing Terminals 500 coverage area

Claims

1. A method for managing drawing data, including raster data, The steps include: generating vector data by vectorizing the aforementioned drawing data, The steps include: identifying first and second nodes in the vector data, thereby generating dimension line data associated with the identified first and second nodes; The steps include identifying one or more connected edges adjacent to the dimension line represented by the dimension line data from the vector data, The steps include identifying a region containing one or more connected edges within the vector data, and performing character recognition on the region in the drawing data corresponding to the identified region, The steps include: storing the characters obtained by the character recognition process as dimension values ​​in association with the dimension line data; A method that includes this.

2. The method according to claim 1, The step of generating the dimension line data is: The steps include detecting multiple tip vertex symbols from the aforementioned drawing data, The steps include determining at least one pair of tip vertex symbols among the detected plurality of tip vertex symbols, The steps include identifying nodes that are adjacent to or closest to the coordinates of the vector data corresponding to the first and second tip vertices of the pair of tip vertex symbols as the first and second nodes in the vector data, respectively. A method that includes this.

3. The method according to claim 1 or 2, A method further comprising the step of determining the direction of the longer side of the identified region, or the region in the drawing data corresponding to the identified region, as the orientation for character recognition.

4. A method according to any one of claims 1 to 3, The step of identifying the one or more connected edges is: A method comprising the step of identifying one or more connected edges located within a predetermined range in a direction perpendicular to an edge connecting the first and second nodes, or a line segment extending beyond one of the nodes of the edge.

5. A method according to any one of claims 1 to 4, A method further comprising the step of performing a thinning process on the drawing data before the vectorization.

6. A program for causing a computer to perform a method for managing drawing data including raster data, wherein the method is The steps include: generating vector data by vectorizing the aforementioned drawing data, The steps include: identifying first and second nodes in the vector data, thereby generating dimension line data associated with the identified first and second nodes; The steps include identifying one or more connected edges adjacent to the dimension line represented by the dimension line data from the vector data, The steps include identifying a region containing one or more connected edges within the vector data, and performing character recognition on the region in the drawing data corresponding to the identified region, The steps include: storing the characters obtained by the character recognition process as dimension values ​​in association with the dimension line data; A program that includes this.

7. A device for managing drawing data, including raster data, The aforementioned drawing data is vectorized to generate vector data, By identifying the first and second nodes in the vector data, dimension line data associated with the identified first and second nodes is generated. From the vector data, one or more connected edges adjacent to the dimension line represented by the dimension line data are identified. The region containing the one or more identified connected edges is identified within the vector data, and character recognition is performed on the region in the drawing data corresponding to the identified region. A device that stores characters obtained by the aforementioned character recognition as dimension values ​​in association with the dimension line data.

Images