Drawing data processing device, drawing data processing system, drawing data processing method, and program

Abstract

This drawing data processing device includes: an acquisition unit that acquires drawing data including a drawing image; an extraction unit that extracts a component table image from the drawing data; a structure recognition unit that recognizes a table structure regarding the component table image; and a character recognition unit that recognizes a character string included in the component table image. With regard to the component table image, by integrating a result of recognizing the table structure with a result of recognizing the character string, component table data obtained by converting information presented in the component table image into data is generated.

Description

Drawing data processing device, drawing data processing system, drawing data processing method and program

[0001] The disclosed technology relates to a drawing data processing device, a drawing data processing system, a drawing data processing method, and a program.

[0002] The following techniques are known as techniques related to processing drawing data. For example, Patent Document 1 describes an image processing device that includes a table recognition means that extracts a parts list that is described in a design drawing of a product and performs character recognition on the names of the listed parts, a vector modeling means that converts each part in the parts list that is recognized by the recognition means and a product identified in the design drawing into a vector model, and a vector model registration means that registers the vectorized parts and products that have been vectorized in association with the design drawing.

[0003] Japanese Patent Application Laid-Open No. 2018-032225

[0004] For example, product drawings used in the manufacturing industry may include a bill of materials (BOM), which is a table containing information about the multiple parts that make up the product. The BOM included in the product drawing is recorded as an image, and the information indicated by the BOM is not digitized. Therefore, a system cannot recognize the information indicated by the BOM included in the product drawing. For example, it is not possible to access data related to each part that makes up the product (e.g., the part's drawing, master information, and transaction information) from the product drawing, or to perform data searches using the information included in the BOM. The above operations would be possible if the information indicated by the BOM, including the table structure, were digitized. While it is possible to manually digitize the information indicated by the BOM, this is extremely cumbersome and it is difficult to ensure the accuracy of the data.

[0005] The disclosed technology has been made in consideration of the above points, and aims to realize, through computer processing, the digitization of information indicated by a parts table image included in drawing data.

[0006] The drawing data processing device according to the disclosed technique includes an acquisition unit that acquires drawing data including drawing images, an extraction unit that extracts a parts table image from the drawing data, a structure recognition unit that recognizes a table structure of the parts table image, and a character recognition unit that recognizes character strings included in the parts table image. The device generates parts table data by digitizing information indicated by the parts table image by integrating the results of the table structure recognition and the character string recognition for the parts table image.

[0007] The structure recognition unit may extract a plurality of cells constituting a parts table from the parts table image, and identify, for each of the extracted cells, whether the cell is a header cell in which a character string indicating an item type is stored or a body cell in which a value for each item type is stored.

[0008] The structure recognition unit may identify a correspondence between the header cell and the body cell based on a positional relationship between the header cell and the body cell.

[0009] The bill of materials data may be generated by assigning character strings recognized by the character recognition unit to the header cells and the body cells whose correspondences have been identified by the structure recognition unit.

[0010] The extraction unit may extract a table image from the drawing data, determine whether the extracted table image is a parts table image, and extract the table image determined to be a parts table image as the parts table image.

[0011] The drawing data processing device may further include a recording processing unit that records the drawing data and the parts list data in association with each other.

[0012] The drawing data processing device may further include a display processing unit that, when an access request for the drawing data is made, performs processing to display, side by side on a display screen, an image contained in the drawing data related to the access request and the contents of the bill of materials data generated based on the drawing data related to the access request.

[0013] The drawing data processing device may further include a link generating unit that embeds a hyperlink indicating a storage location of data related to a part in a character string indicating identification information of the part included in the parts table data.

[0014] A drawing data processing system according to the disclosed technique includes the drawing data processing device described above and a terminal device communicably connected to the drawing data processing device.

[0015] The drawing data processing method according to the disclosed technology involves a computer executing a process to acquire drawing data including a drawing image, extract a parts table image from the drawing data, recognize a table structure for the parts table image, recognize character strings included in the parts table image, and generate parts table data in which information indicated by the parts table image is digitized by integrating the results of recognizing the table structure and the results of recognizing the character strings for the parts table image.

[0016] The program related to the disclosed technology is a program for causing a computer to execute a process of acquiring drawing data including a drawing image, extracting a parts table image from the drawing data, recognizing a table structure for the parts table image, recognizing character strings included in the parts table image, and integrating the results of recognizing the table structure and the results of recognizing the character strings for the parts table image to generate parts table data that digitizes information indicated by the parts table image.

[0017] According to the disclosed technology, it is possible to realize the digitization of information indicated by a parts table image included in drawing data by computer processing.

[0018] 1 is a diagram showing an example of the configuration of a drawing data processing system according to an embodiment of the disclosed technology. FIG. 2 is a diagram showing an example of the contents of drawing data according to an embodiment of the disclosed technology. FIG. 3 is a diagram showing an example of the hardware configuration of a drawing data processing device according to an embodiment of the disclosed technology. FIG. 4 is a functional block diagram showing an example of the functional configuration of a drawing data processing device according to an embodiment of the disclosed technology. FIG. 5 is a diagram showing an example of a mode of recognizing the table structure of a parts bill of materials image according to an embodiment of the disclosed technology. FIG. 6 is a diagram showing an example of a mode of recognizing the table structure of a parts bill of materials image according to an embodiment of the disclosed technology. FIG. 7 is a diagram showing an example of a mode of assigning recognized character strings to header cells and body cells according to an embodiment of the disclosed technology. FIG. 8 is a diagram showing an example of a display screen of a terminal device according to an embodiment of the disclosed technology. FIG. 9 is a flowchart showing an example of the flow of processing performed in a drawing data processing device according to an embodiment of the disclosed technology.

[0019] Hereinafter, embodiments of the disclosed technology will be described with reference to the drawings. In each drawing, substantially the same or equivalent components or parts are denoted by the same reference numerals.

[0020] 1 is a diagram illustrating an example of the configuration of a drawing data processing system 1 according to an embodiment of the disclosed technology. The drawing data processing system 1 includes a drawing data processing device 10 and a plurality of terminal devices 20. The drawing data processing device 10 and each of the plurality of terminal devices 20 are communicably connected via a network 30 such as the Internet.

[0021] Each of the terminal devices 20 can make a request to register drawing data or a request to access drawing data to the drawing data processing device 10. The terminal devices 20 can access the drawing data processing device 10, for example, through a website or a dedicated application. When the terminal device 20 performs a registration process for drawing data through the website or the dedicated application, the drawing data is transmitted to the drawing data processing device 10. The drawing data processing device 10 has a function of generating bill of materials data by digitizing information indicated by a bill of materials image included in the drawing data.

[0022] 2 is a diagram showing an example of the contents of drawing data to be processed by the drawing data processing device 10. The drawing data 200 is image data including a drawing image of a product or part, and has a file format such as JPEG, PNG, GIF, or PDF. FIG. 2 shows an example of drawing data 200 including a drawing image 201 of a metal processed product consisting of multiple parts. In addition to the drawing image 201, the drawing data 200 also includes a title block image 202 and a parts list image 203. The parts list image 203 is a table showing information such as the reference numbers, quantities, part numbers, and materials of the parts that make up the product shown in the drawing image 201.

[0023] 3 is a diagram showing an example of the hardware configuration of the drawing data processing device 10. The drawing data processing device 10 is a computer, and includes a CPU (Central Processing Unit) 101, a RAM (Random Access Memory) 102, a non-volatile memory 103, an input / output interface (I / F) 104, a network interface (I / F) 105, and a read / write unit 106. These pieces of hardware are connected to each other via a bus 107.

[0024] The non-volatile memory 103 is a non-volatile storage medium such as a hard disk or flash memory. An image data processing program 110 is stored in the non-volatile memory 103. The RAM 102 is a work memory for the CPU 101 to execute processing. The CPU 101 loads the drawing data processing program 110 into the RAM 102 and executes processing in accordance with the drawing data processing program 110. Input / output devices (not shown) such as a mouse, keyboard, and display are connected to the input / output interface 104. The network interface 105 is an interface for connecting the drawing data processing device 10 to the network 30. The read / write unit 106 controls reading and writing of data from and to the non-volatile memory 103.

[0025] 4 is a functional block diagram showing an example of the functional configuration of the drawing data processing device 10. The drawing data processing device 10 has an acquisition unit 11, an extraction unit 12, a structure recognition unit 13, a character recognition unit 14, a link generation unit 15, a recording processing unit 16, and a display processing unit 17. When the CPU 101 executes the drawing data processing program 110, the drawing data processing device 10 functions as the acquisition unit 11, the extraction unit 12, the structure recognition unit 13, the character recognition unit 14, the link generation unit 15, the recording processing unit 16, and the display processing unit 17.

[0026] The acquisition unit 11 acquires drawing data transmitted from the terminal device 20. The extraction unit 12 extracts a parts table image from the drawing data acquired by the acquisition unit 11. Specifically, the extraction unit 12 extracts a parts table image from the drawing data by performing two-stage processing, a first processing and a second processing, which will be described below.

[0027] The first process is a process of extracting all images expressed in a table format including ruled lines from the drawing data as table images. The extraction unit 12 performs the first process using an extraction model constructed by machine learning that extracts table images as target objects from the input image.

[0028] The second process is a process for determining whether the table image extracted by the first process is a BOM image. The extraction unit 12 performs the second process using a determination model constructed by machine learning. A BOM image lists information about multiple parts by item and has characteristics different from other table images (e.g., the title block image 202) included in the drawing data. Therefore, by using a trained determination model, it is possible to determine whether a table image is a BOM image. The machine learning of the determination model can be performed, for example, using multiple training data consisting of pairs of BOM images and table images other than the BOM image, and correct labels indicating whether these table images are BOM images. The extraction unit 12 extracts, from the table images extracted in the first process, those determined to be BOM images as BOM images. For example, for drawing data 200 illustrated in FIG. 2 , the extraction unit 12 extracts the title block image 202 and the BOM image 203 in the first process, and extracts the BOM image 203 in the second process.

[0029] The structure recognition unit 13 recognizes the table structure of the BOM image extracted by the extraction unit 12. FIG. 5 is a diagram showing an example of how the structure recognition unit 13 recognizes the table structure of the BOM image. First, the structure recognition unit 13 extracts a plurality of cells constituting the BOM from the BOM image. The extraction unit 12 extracts, as one cell, an area surrounded by a rectangular frame formed by tracing ruled lines included in the BOM image. FIG. 5 shows a plurality of cells 210 extracted from the BOM image 203.

[0030] Next, the structure recognition unit 13 identifies whether each of the multiple cells extracted from the BOM image is a header cell or a body cell. A header cell is a cell that stores a string indicating an item type in the BOM, such as "Reference Number," "Quantity," "Part Number," or "Material." A body cell is a cell that stores a value for each item type, such as "1," "20220929-BVC-31," or "SS400." The structure recognition unit 13 identifies cells using an identification model constructed by machine learning. The machine learning of the identification model can be performed using, for example, multiple training data sets consisting of pairs of BOM images in which cells are identified and correct labels indicating whether each cell in the BOM image is a header cell or a body cell. Figure 5 illustrates multiple cells 210 that have been identified as header cells or body cells.

[0031] The structure recognition unit 13 also identifies the correspondence between header cells and body cells based on their positional relationship. Specifically, the structure recognition unit 13 identifies a body cell that is in a specific positional relationship with a header cell as a body cell that belongs to that header cell. For example, as shown in FIG. 6, for a header cell located in the first column and first row, the structure recognition unit 13 identifies each of the body cells located in the second to sixth rows of the first column as a body cell that belongs to that header cell. Similarly, for a header cell located in the first column and first row, the structure recognition unit 13 identifies each of the body cells located in the second to sixth rows of the second column as a body cell that belongs to that header cell. Similar identification is performed for each cell in the third and fourth columns.

[0032] The structure recognition unit 13 also identifies the correspondence between body cells based on the positional relationship between the header cells and the body cells. Specifically, the structure recognition unit 13 identifies multiple body cells aligned in the same direction as multiple header cells as a group of cells belonging to the same part. For example, as shown in Figure 7, the structure recognition unit 13 identifies each body cell located in the same row and column as a group of cells belonging to the same part. The structure recognition unit 13 performs the above processes to recognize the table structure of the BOM image.

[0033] The character recognition unit 14 recognizes character strings included in the BOM image extracted by the extraction unit 12. Specifically, the character recognition unit 14 performs character recognition on each cell extracted from the BOM image by the structure recognition unit 13. The character recognition by the character recognition unit 14 can be performed using a known OCR (Optical Character Reader) technology. The character recognition unit 14 assigns the recognized character strings to each of the header cells and body cells whose correspondences have been identified by the structure recognition unit 13. For example, as shown in FIG. 8 , the character recognition unit 14 assigns the recognized character strings "Reference Number," "Quantity," "Part Number," and "Material" to four header cells, respectively. Furthermore, the character recognition unit 14 assigns the recognized character strings "SS400," "A5052," "A7076," "A5052," and "A6063" to five body cells belonging to the header cell to which the character string "Material" is assigned.

[0034] In this way, the results of recognizing the structure of the BOM image and the results of recognizing the character strings are integrated. As a result, BOM data is generated in which the information indicated by the BOM image, including the table structure, is digitized. The BOM data may have a format such as CSV (Comma Separated Values) data.

[0035] The link generation unit 15 embeds a hyperlink indicating the storage location of data related to a part in a character string indicating the identification information (e.g., part number or drawing number) of the part included in the BOM data. "Data related to the part" includes, for example, drawing data, master information, and transaction information for the part. The BOM data includes identification information for each of multiple parts, such as the part number or drawing number. The link generation unit 15 determines whether data related to the part exists in the system based on the part identification information included in the BOM data. The link generation unit 15 makes this determination by searching, for example, the non-volatile memory 103 or an external storage (not shown). If the link generation unit 15 determines that data related to the part exists in the system, it embeds a hyperlink indicating the storage location of the data in a character string indicating the identification information of the part included in the BOM data.

[0036] The recording processing unit 16 records the drawing data and the parts list data generated based on the drawing data in association with each other in the non-volatile memory 103 .

[0037] When an access request for drawing data is made, the display processing unit 17 performs processing to display the contents of the drawing data related to the access request and the contents of the bill of materials data generated based on the drawing data related to the access request side by side on the display screen. The access request for drawing data is issued from the terminal device 20. Based on a user operation, the terminal device 20 transmits the access request for the drawing data to the drawing data processing device 10 together with identification information of the drawing (e.g., a part number or drawing number).

[0038] When the display processing unit 17 receives an access request, it reads out the drawing data related to the access request together with the corresponding parts list data from the non-volatile memory 103, and generates display data for displaying these contents on the display screen of the terminal device 20 that issued the access request. The display processing unit 17 transmits the generated display data to the terminal device 20 that issued the access request.

[0039] 9 is a diagram illustrating an example of a display screen of the terminal device 20 that has received display data. The display screen 21 of the terminal device 20 displays an image 300 (including a drawing image, a title block image, and a parts list image) included in the drawing data related to the access request, alongside a parts list 310, which is the contents of the parts list data generated based on the drawing data related to the access request. A hyperlink indicating the storage location of data related to the part (e.g., the drawing data, master information, and transaction information for the part) is embedded in a character string indicating the part number, which is the identification information for each part included in the parts list 310. A user of the terminal device 20 can access data related to the part (e.g., the drawing data, master information, and transaction information for the part) by clicking the character string indicating the part number using the mouse pointer 320. For example, the drawing data for the part may be displayed as a pop-up on the display screen of the terminal device 20.

[0040] 10 is a flowchart showing an example of the flow of processing performed by the CPU 101 of the drawing data processing device 10 when the CPU 101 executes the drawing data processing program 110. For example, when a user operates the terminal device 20 to perform a registration process of drawing data, the drawing data processing program 110 is executed.

[0041] In step S1, the acquisition unit 11 acquires drawing data transmitted from the terminal device 20. In step S2, the extraction unit 12 extracts a parts table image from the drawing data acquired in step S2. Specifically, the extraction unit 12 performs a first process to extract all images represented in a table format including ruled lines from the drawing data as table images. Subsequently, the extraction unit 12 performs a second process to determine whether the table images extracted by the first process are parts table images. The extraction unit 12 extracts the table images determined to be parts table images as parts table images. Note that if a table image cannot be extracted from the drawing data or if there is no table image determined to be a parts table image, this routine ends.

[0042] In step S3, the structure recognition unit 13 recognizes the table structure of the BOM image extracted in step S2. Specifically, the structure recognition unit 13 extracts multiple cells that make up the BOM from the BOM image. Next, the structure recognition unit 13 identifies whether each of the multiple extracted cells is a header cell or a body cell. Next, the structure recognition unit 13 identifies the correspondence between the header cell and the body cell based on the positional relationship between the header cell and the body cell. Furthermore, the structure recognition unit 13 identifies the correspondence between the body cells based on the positional relationship between the header cell and the body cell.

[0043] In step S4, the character recognition unit 14 performs character recognition on each cell extracted from the BOM image in step S3. The character recognition unit 14 assigns the recognized character string to each of the header cells and body cells whose correspondence has been identified by the structure recognition unit 13. In other words, the results of recognizing the structure and the results of recognizing the character string for the BOM image are integrated. As a result, BOM data is generated in which the information indicated by the BOM image, including the table structure, is digitized.

[0044] In step S5, it is determined whether data related to the part (drawing data, master information, and transaction information for the part) exists in the system based on the part identification information included in the parts bill data generated in step S4. If it is determined that data related to the part exists, the process proceeds to step S6, and if it is determined that data related to the part does not exist, the process proceeds to step S7.

[0045] In step S6, the link generation unit 15 embeds a hyperlink indicating the storage location of data related to the part in a string indicating the identification information (e.g., product number or drawing number) of the part included in the parts table data generated in step S4.

[0046] In step S7, the recording processing unit 16 records in the non-volatile memory 103 the drawing data acquired in step S1 and the parts list data into which the hyperlink has been embedded in step S6 in association with each other.

[0047] In step S8, the display processing unit 17 determines whether or not there is a request to access the drawing data. The request to access the drawing data is issued from the terminal device 20. If it is determined that there is a request to access the drawing data, the process proceeds to step S9.

[0048] In step S9, the display processing unit 17 performs processing to display the contents of the drawing data related to the access request and the contents of the parts list data generated based on the drawing data related to the access request side by side on the display screen of the terminal device 20 that issued the access request.

[0049] As described above, the drawing data processing device 10 according to the embodiment of the disclosed technology includes an acquisition unit 11 that acquires drawing data including drawing images, an extraction unit 12 that extracts a parts table image from the drawing data, a structure recognition unit 13 that recognizes a table structure for the extracted parts table image, and a character recognition unit 14 that recognizes character strings included in the extracted parts table image. The drawing data processing device 10 generates parts table data in which information indicated by the parts table image, including the table structure, is digitized by integrating the results of the structure recognition and the character string recognition for the parts table image.

[0050] For example, product drawings used in the manufacturing industry may contain a bill of materials (BOM), which is a table that displays information about the multiple components that make up the product. The BOM included in a product drawing is recorded as an image, and the information it represents is not digitized. Therefore, a system cannot recognize the information represented by the BOM from the BOM included in the product drawing. For example, it is not possible to use a product drawing as a starting point to access data related to each component that makes up the product (e.g., the component's drawing, master information, and transaction information), or to perform an information search using a part number or other character string listed in the BOM as a search term. The above operations would be possible if the information represented by the BOM, including the table structure, were digitized. While it is possible to manually digitize the information represented by the BOM, this is extremely cumbersome and it is difficult to ensure the accuracy of the data.

[0051] The drawing data processing device 10 makes it possible to digitize information indicated by a parts table image included in drawing data through computer processing. Furthermore, the drawing data processing device 10 generates parts table data in which the information indicated by the parts table image included in the drawing data is digitized, including the table structure. This enables the system to recognize the information indicated by the parts table image. As a result, for example, operations such as accessing data related to each part that constitutes the product (e.g., the part's drawing, master information, and transaction information) starting from the drawing data of a product drawing, or performing data searches using information included in the parts table, become possible.

[0052] In the above description, character recognition is performed after table structure recognition is performed on a BOM image, but the disclosed technology is not limited to this example. Character recognition may be performed on a BOM image before table structure recognition. In this case, header cells and body cells may be identified using the character string recognition results.

[0053] In the above description, the header cell / body cell identification is performed using an identification model constructed by machine learning, but the disclosed technology is not limited to this example. The header cell / body cell identification may also be performed based on rules. For example, the cells located at the top (each column of the first row) may be identified as header cells, and the other cells may be identified as body cells.

[0054] In the above description, a case where a BOM image is extracted from drawing data using an extraction model and a judgment model constructed by machine learning has been exemplified, but the disclosed technology is not limited to this example. A BOM image may also be extracted from drawing data using a rule base. For example, objects located in a specific range in a drawing may be extracted as a BOM image.

[0055] The following supplementary note is further disclosed regarding the above embodiment: (Supplementary note 1) A drawing data processing device including: an acquisition unit that acquires drawing data including a drawing image, an extraction unit that extracts a parts table image from the drawing data, a structure recognition unit that recognizes a table structure of the parts table image, and a character recognition unit that recognizes character strings included in the parts table image, and which generates parts table data in which information indicated by the parts table image is digitized by integrating a result of the table structure recognition and a result of the character string recognition for the parts table image.

[0056] (Supplementary Note 2) The drawing data processing device according to Supplementary Note 1, wherein the structure recognition unit extracts a plurality of cells constituting a parts table from the parts table image, and identifies, for each of the extracted cells, whether the cell is a header cell in which a character string indicating an item type is stored or a body cell in which a value for each item type is stored.

[0057] (Supplementary Note 3) The drawing data processing device according to Supplementary Note 2, wherein the structure recognition unit identifies a correspondence between the header cell and the body cell based on a positional relationship between the header cell and the body cell.

[0058] (Supplementary Note 4) The drawing data processing device according to Supplementary Note 3, wherein the bill of materials data is generated by assigning a character string recognized by the character recognition unit to each of the header cells and the body cells whose correspondence has been identified by the structure recognition unit.

[0059] (Supplementary Note 5) The drawing data processing device according to any one of Supplementary Note 1 to Supplementary Note 4, wherein the extraction unit extracts a table image from the drawing data, determines whether the extracted table image is a parts table image, and extracts the table image determined to be a parts table image as the parts table image.

[0060] (Supplementary Note 6) The drawing data processing device according to any one of Supplementary Note 1 to Supplementary Note 5, further comprising a recording processing unit that records the drawing data and the parts list data in association with each other.

[0061] (Supplementary Note 7) A drawing data processing device according to any one of Supplementary Notes 1 to 6, further comprising a display processing unit that, when an access request for the drawing data is made, performs processing to display, side by side on a display screen, an image included in the drawing data related to the access request and the contents of the bill of materials data generated based on the drawing data related to the access request.

[0062] (Supplementary Note 8) The drawing data processing device according to any one of Supplementary Note 1 to Supplementary Note 7, further comprising a link generation unit that embeds a hyperlink indicating a storage location of data related to a part in a character string indicating identification information of the part included in the parts table data.

[0063] (Supplementary Note 9) A drawing data processing system including: the drawing data processing device according to any one of Supplementary Notes 1 to 8; and a terminal device communicably connected to the drawing data processing device.

[0064] (Supplementary Note 10) A drawing data processing method in which a computer executes the following processes: acquire drawing data including a drawing image; extract a parts table image from the drawing data; recognize a table structure for the parts table image; recognize character strings included in the parts table image; and generate parts table data in which information indicated by the parts table image is digitized by integrating the results of the table structure recognition and the results of the character string recognition for the parts table image.

[0065] (Supplementary Note 11) A program for causing a computer to execute the following processes: acquiring drawing data including a drawing image; extracting a parts table image from the drawing data; recognizing a table structure for the parts table image; recognizing character strings included in the parts table image; and generating parts table data in which information indicated by the parts table image is digitized by integrating the results of recognizing the table structure and the results of recognizing the character strings for the parts table image.

Claims

1. A drawing data processing device comprising: an acquisition unit that acquires drawing data including a drawing image; an extraction unit that extracts a parts table image from the drawing data; a structure recognition unit that recognizes a table structure for the parts table image; and a character recognition unit that recognizes character strings included in the parts table image, and which generates parts table data in which information indicated by the parts table image is digitized by integrating the results of the table structure recognition and the character string recognition for the parts table image.

2. The drawing data processing device according to claim 1, wherein the structure recognition unit extracts a plurality of cells constituting a parts table from the parts table image, and identifies, for each of the extracted cells, whether the cell is a header cell in which a character string indicating an item type is stored or a body cell in which a value for each item type is stored.

3. The drawing data processing device according to claim 2, wherein the structure recognition unit identifies the correspondence between the header cell and the body cell based on the positional relationship between the header cell and the body cell.

4. The drawing data processing device according to claim 3, wherein the bill of materials data is generated by assigning a character string recognized by the character recognition unit to each of the header cells and body cells whose correspondence has been identified by the structure recognition unit.

5. The drawing data processing device according to claim 1, wherein the extraction unit extracts a table image from the drawing data, determines whether the extracted table image is a parts table image, and extracts the table image determined to be a parts table image as the parts table image.

6. The drawing data processing device according to claim 1, further comprising a recording processing unit that records the drawing data and the parts list data in association with each other.

7. A drawing data processing device as claimed in claim 1, further comprising a display processing unit that, when an access request for the drawing data is made, performs processing to display, side by side on a display screen, an image contained in the drawing data related to the access request and the contents of the parts list data generated based on the drawing data related to the access request.

8. The drawing data processing device according to claim 1, further comprising a link generating section for embedding a hyperlink indicating a storage location of data related to a part in a character string indicating identification information of the part included in the parts table data.

9. A drawing data processing system comprising: a drawing data processing device according to any one of claims 1 to 8; and a terminal device communicably connected to said drawing data processing device.

10. A drawing data processing method in which a computer executes the following processes: acquire drawing data including a drawing image; extract a parts table image from the drawing data; recognize a table structure for the parts table image; recognize character strings included in the parts table image; and generate parts table data in which information indicated by the parts table image is digitized by integrating the results of recognizing the table structure and the results of recognizing the character strings for the parts table image.

11. A program for causing a computer to execute the following processes: acquiring drawing data including a drawing image; extracting a parts list image from the drawing data; recognizing a table structure for the parts list image; recognizing character strings included in the parts list image; and integrating the results of recognizing the table structure and the results of recognizing character strings for the parts list image to generate parts list data that digitizes information indicated by the parts list image.

Images