Information processing device, information processing method, and information processing program

Abstract

This information processing device accepts the designation of unit system data representing a unit system, the designation of numerical data compliant with the unit system data, and the designation of attribute data representing attributes of the numerical data. The information processing device outputs target image data representing a drawing of a target object and satisfying conditions represented by the unit system data, the numerical data, and the attribute data.

Description

Information processing apparatus, information processing method, and information processing program

[0001] The present disclosure relates to an information processing apparatus, an information processing method, and an information processing program.

[0002] Conventionally, there has been known a CAD system that quickly and efficiently converts dimension indications in technical drawings (for example, Japanese Patent No. 7413445). This CAD system converts dimension indications of technical drawings from a first dimension indication standard having a first dimension display attribute to a second dimension indication standard having a second dimension display attribute (Japanese Unexamined Patent Publication No. 8-287138).

[0003] There is also known a drawing creation system for other countries that can automatically rewrite drawings so as to conform to the language, standards, and the like of a specified country (Japanese Unexamined Patent Publication No. 6-259487). This drawing creation system for other countries sets, via an operation panel, conditions including the character language, metrological unit system, drawing standard, and the like of source and target drawings. Then, the drawing creation system for other countries reads a drawing to be converted with a scanner. Further, when drawing information is stored in an optical disc, the drawing creation system for other countries calls the drawing by specifying a drawing number. The drawing creation system for other countries separates the input information into text data and graphic data, and further separates the text data into numerical data and character data. Among these, the numerical data is converted and changed when there is a change in the metrological system. For example, 25.4 mm is replaced with 1 inch. In the drawing creation system for other countries, character information is translated, and graphic data is converted so as to conform to the standards of the target country. The drawing creation system for other countries integrates these pieces of data and outputs them as a drawing for the target country.

[0004] By the way, in Japan, Europe and other regions, the metric system is the mainstream unit system. Therefore, in Japan, Europe and other regions, various dimensions described on drawings are often described based on millimeters. On the other hand, in the United States of America, the United Kingdom and other countries, the yard-pound system is the mainstream unit system. Therefore, in the United States of America, the United Kingdom and other countries, various dimensions described on drawings are often described based on inches or feet.

[0005] Users handling drawings may want to retrieve a drawing that meets their desired numerical conditions from among multiple drawings conforming to different unit systems, as described above. For example, a user may want to retrieve a drawing from among several drawings that include drawings with dimensions in millimeters and drawings with dimensions in inches or feet, where the dimensions of a specific part of the object depicted in the drawing are within a specified numerical range.

[0006] However, drawings often lack direct indication of the unit system they adhere to, and unlike CAD data, typical drawing file formats such as PDF data cannot internally store unit system data.

[0007] Therefore, even if one were to store a large amount of image data in a database and try to find a drawing in which the dimensions of a specific part of an object are within a specified numerical range, it would be extremely difficult to obtain image data in which the object depicted in the drawing meets the specified dimensional conditions, because the image data does not include a unit system description, and multiple unit systems are mixed together.

[0008] The CAD system described in Patent Document 1 assumes that the CAD data internally holds unit system data. However, for example, PDF drawing data, which is an example of raster data format, often does not contain a description of the unit system. Therefore, the CAD system in Patent Document 1 cannot determine the meaning of the read numbers or determine the unit system simply by reading the file data.

[0009] Furthermore, the drawing creation system for other countries described in Patent Document 2 requires a means to specify the unit systems before and after conversion. Also, the drawing creation system for other countries described in Patent Document 2 is a system for creating drawings and cannot perform cross-sectional extraction or comparison of information related to unit systems.

[0010] This disclosure is made in light of the circumstances described above, and aims to obtain image data from among image data representing drawings created in accordance with different unit systems, in which the objects depicted in the drawings satisfy predetermined numerical conditions.

[0011] To achieve the above objective, a first aspect of this disclosure is an information processing device comprising: a receiving unit that receives the specification of unit system data representing a unit system, the specification of numerical data compliant with the unit system data, and the specification of attribute data representing the attributes of the numerical data; and an output unit that outputs target image data representing a drawing of an object, and that satisfies the conditions represented by the unit system data and the numerical data.

[0012] Furthermore, a second aspect of this disclosure is an information processing method in which a computer receives the specification of unit system data representing a unit system, the specification of numerical data compliant with the unit system data, and the specification of attribute data, and outputs target image data representing a drawing of an object, which satisfies the conditions represented by the unit system data and the numerical data.

[0013] Furthermore, a third aspect of this disclosure is an information processing program for causing a computer to perform processing that accepts the specification of unit system data representing a unit system, the specification of numerical data compliant with the unit system data, and the specification of attribute data, and outputs target image data representing a drawing of an object, and that satisfies the conditions represented by the unit system data and the numerical data.

[0014] According to this disclosure, the effect is obtained that, from image data representing drawings created in accordance with different unit systems, image data in which the objects drawn in the drawings satisfy predetermined numerical conditions can be obtained.

[0015] This figure shows an example of the schematic configuration of the information processing system of this embodiment. This figure shows an example of a trained model used in this embodiment. This figure shows an example of related data used in this embodiment. This figure shows an example of image data. This is an example of a screen displayed on the display unit of a user terminal. This is an example of a screen displayed on the display unit of a user terminal. This is a schematic block diagram of a computer that functions as each device of the information processing system. This figure illustrates the processing performed by the information processing system of this embodiment. This figure illustrates the processing performed by the information processing system of this embodiment.

[0016] The embodiments will be described in detail below with reference to the drawings.

[0017] <System Configuration of the Information Processing System> Figure 1 is a block diagram of the information processing system 10 of this embodiment. As shown in Figure 1, the information processing system 10 of this embodiment includes an information processing device 16 and a plurality of user terminals 18A, 18B, 18C. In the following description, unless a specific terminal is being referred to, one user terminal will be referred to as user terminal 18. The information processing device 16 and the user terminals 18 are connected to each other via a network 19, such as the Internet.

[0018] The information processing system 10 of this embodiment is a system for obtaining image data that satisfies predetermined attribute numerical conditions from among multiple image data representing drawings created in accordance with different unit systems.

[0019] (Information Processing Device 16) The information processing device 16 is a server that responds to information transmitted from the user terminal 18. As shown in Figure 1, the information processing device 16 functionally comprises a reception unit 20, a creation unit 22, a search unit 24, an output unit 26, a trained model storage unit 28, an image database 30, and a related data storage unit 32. The image database 30 is an example of a database in this disclosure.

[0020] The information processing device 16 of this embodiment uses a trained model, described later, to recognize various information within multiple image data stored in the image database 30, and based on the recognition results, creates related data (table-format data) for searching multiple image data stored in the image database 30. Then, when the information processing device 16 searches for image data that satisfies the search conditions from among the multiple image data stored in the image database 30, it refers to the related data to search for the image data. This will be explained in detail below.

[0021] The trained model storage unit 28 stores trained models for recognizing information within image data. Specifically, the trained model storage unit 28 stores a trained model for unit system discrimination, a trained model for dimension recognition, a trained model for dimension type identification (which recognizes dimension types as an example of attributes), and a trained model for character string image recognition. These trained models are examples of image recognition means. Furthermore, these trained models are models that have been pre-constructed using known artificial intelligence techniques and known machine learning techniques. Machine learning models corresponding to trained models include, for example, convolutional neural networks. In this embodiment, the case of recognizing information within image data using a machine learning-based trained model will be explained as an example, but it is also possible to use a combination of rule-based methods instead of using only trained models. Figure 2 is a diagram illustrating the trained model for unit system discrimination, the trained model for dimension recognition, the trained model for dimension type identification, and the trained model for character string image recognition. Note that dimensions are an example of numerical attributes described in the drawing, and numerical values ​​corresponding to other attributes can be recognized in the same way.

[0022] As shown in Figure 2(A), the trained model for unit system discrimination is a model that, upon input of image data, outputs a result indicating the determination of the unit system to which the drawing conforms. Specifically, as shown in Figure 2(A), when image data representing a drawing is input to the trained model for unit system discrimination, information is output regarding whether the drawing conforms to inches or millimeters. By using a machine learning model for unit system discrimination, it becomes possible to perform unit system discrimination not only for drawings with a limited number of layouts and formats, but also for drawings with different layouts and formats depending on the company or organization.

[0023] As shown in Figure 2(B), the trained model for dimension recognition is a model that, when drawing data is input, detects the area on the drawing where the dimension information of the object is described (for example, the vertex positions of the rectangular area where the dimensions are described). Specifically, as shown in Figure 2(B), when image data is input to the trained model for dimension recognition, it detects the broad-sense dimension information of the image data. The broad-sense dimension information refers to the rectangular area on the drawing that contains information centered on the dimension values, and specifically, as in the example in Figure 2, it may be a string of only numbers such as "266" or "241", or it may be a string of letters and numbers, such as "φ0.197" as in Figure 1, which has the Greek letter φ indicating diameter attached. On the other hand, dimensional information in the narrow sense refers to numerical information that includes information about the type of dimension, such as "the sheet thickness of the raw material for manufacturing the target product, which is a sheet metal product, is 0.079 inches," or "the dimensions of the x, y, and z axes of the completed target product are 21.181 × 32.992 × 0.079 inches, respectively." This is the information that we want to obtain through a series of processes, including subsequent processing. The significance of using a machine learning model is the same as that of a trained model for unit system discrimination, and the significance of using the various machine learning models described later is also the same.

[0024] As shown in Figure 2(C), the pre-trained model for dimension type identification is a pre-trained model that, when broad dimension information recognized by the pre-trained model for dimension recognition is input, outputs dimension type information that represents the type of dimension, which is a type of attribute. Specifically, as shown in Figure 2(C), when broad dimension information recognized by the pre-trained model for dimension recognition is input to the pre-trained model for dimension type identification, the identification result of the dimension type information is output. Therefore, the pre-trained model for dimension type identification identifies what type of dimension is included in the broad dimension information mentioned above. For example, it may recognize that the dimension information corresponds to thickness or that the dimension information corresponds to outer diameter.

[0025] As shown in Figure 2(D), the trained model for character recognition is a model that, when given broad-sense dimension information recognized by the trained model for dimension recognition, outputs the recognition result of the characters contained in the broad-sense dimension information. As a result, character images contained in the broad-sense dimension information are recognized as character data containing numbers. For example, the trained model for character recognition recognizes visual notations on a drawing such as "0.079", "4C-0.079", or "6-φ0.197" as actual character data.

[0026] The image database 30 stores image data representing drawings of the object. Typically, the image database 30 stores PDF data of drawings, which is an example of image data representing drawings. In this embodiment, the image data representing drawings of the object is, for example, image data such as a design drawing created using CAD that has been converted into PDF format. The file size of the image data stored in the image database 30 is not a concern. The image data stored in the image database 30 may be resized (reduced) to the minimum size necessary for analysis. In addition, at least some of the image data of the multiple image data stored in the image database 30 do not have the corresponding unit system indicated. By inputting each of these multiple image data into the trained model for unit system discrimination and the trained model for dimension recognition, the unit system and broad-sense dimension information of the image data are identified. Furthermore, by inputting the broad-sense dimension information into the trained model for dimension type identification and the trained model for character image recognition, the dimension type and dimension value of the image data are obtained. The data format of the multiple image data stored in the image database 30 is, for example, at least one of raster data format and vector data format.

[0027] The related data storage unit 32 stores related data for each of the multiple image data stored in the image database 30. Figure 3 is a diagram showing an example of related data. As shown in Figure 3, the related data is associated with dimension No., drawing ID, unit system, dimension type, dimension value, converted unit system, and converted dimension value. The related data shown in Figure 3 shows the details of each data obtained from the image data with drawing ID "dr0000001". For example, dimension No. "0001" is associated with drawing ID "dr0000001", unit system "inch", dimension type "plate thickness", dimension value "0.079", converted unit system "mm", and converted dimension value "2". The unit system "inch" in Figure 3 is the unit system determined by inputting the image data of drawing ID "dr0000001" into the trained model for unit system determination described above. Furthermore, the dimension type "plate thickness" in Figure 3 is a dimension type identified by inputting broad dimension information obtained by inputting the image data of drawing ID "dr0000001" into the aforementioned trained model for dimension recognition into the aforementioned trained model for dimension type identification. Also, the dimension value "0.079" in Figure 3 is a dimension value identified by inputting broad dimension information obtained by inputting the image data of drawing ID "dr0000001" into the aforementioned trained model for dimension recognition into the aforementioned trained model for character image recognition. The converted unit system "mm" in Figure 3 is the unit system to which the unit system "inch" to which the drawing of drawing ID "dr0000001" conforms is converted. The converted dimension value "2" in Figure 3 is the dimension value obtained by converting the dimension value "0.079" in Figure 3, which is in inches, to millimeters. Note that various data about other image data are stored in the part after "..." in the related data of Figure 3.

[0028] The reception unit 20 receives various data transmitted from the user terminal 18. For example, the reception unit 20 receives image data provided by the user terminal 18. Alternatively, for example, the reception unit 20 receives search condition data for searching multiple image data stored in the image database 30.

[0029] Figure 4 shows an example of image data received by the reception unit 20. The image data shown in Figure 4 is drawing DW. T shown in Figure 4 is the title block of drawing DW. Drawing DW in Figure 4 also depicts the object to be processed O. V1 in Figure 4 is a perspective view of object O, V2 is a side view of object O, and V3 is a top view of object O. The dimensions of object O are indicated in the side view V2 and top view V3 of Figure 4. Note that drawing DW in Figure 4 is an example of a drawing in which object O is depicted based on the imperial system, and the dimensions of object O depicted in drawing DW are in inches.

[0030] Users may want to uniformly search for image data that meets their desired search criteria from a database containing a mix of drawings, such as drawing DW shown in Figure 4, which includes drawings where the dimensions of objects are based on inches, drawings where the dimensions are based on millimeters, and drawings where dimensions based on two or more unit systems are listed side by side (for example, drawings where dimensions based on two unit systems are listed in parallel). Specifically, they might want to search for drawings that depict objects with dimensions of 30-50 millimeters in the x, y, and z directions, search for drawings that depict objects with an outer diameter of 3-4 inches, or download the dimensional information from these drawings and import it into a spreadsheet program to analyze material costs or processing costs for a collection of drawings that mix inches and millimeters. However, as mentioned above, image data does not contain data about the unit system, and the unit system is often not indicated on the drawing itself. Therefore, simply performing character recognition and image recognition on the drawing will not provide information about the unit system to which the drawing conforms. Therefore, in this embodiment, various recognition processes are performed using the various pre-trained models described above, and related data, described later, is created based on the recognition results. As mentioned above, the related data, described later, may be created based on recognition results obtained by other image recognition means, not just pre-trained models.

[0031] The creation unit 22 creates the related data stored in the related data storage unit 32. Specifically, the creation unit 22 first inputs the drawing data received by the reception unit 20 into the respective trained models for unit system discrimination, dimension recognition, dimension type identification, and character string image recognition, thereby identifying the unit system, dimension value, and dimension type of the image data.

[0032] Next, the creation unit 22 calculates the converted dimension value by converting the specified dimension value to a dimension value in a different unit system. Then, the creation unit 22 creates related data by associating the drawing ID of the image data, the dimension No., the unit system, dimension value, and dimension type of the image data, the converted unit system, and the converted dimension value.

[0033] For example, when the drawing DW shown in Figure 4 is the target, the creation unit 22 assigns a drawing ID to the drawing DW. Next, the creation unit 22 uses the aforementioned trained models for unit system discrimination, dimension recognition, dimension type identification, and character image recognition to identify the unit system, dimension values, and dimension types of the drawing DW. For example, the plate thickness of the object described in the drawing DW of Figure 4, which is 0.079, is identified, the x-direction dimension of the object in the drawing DW, which is 21.181, is identified, the y-direction dimension of the object in the drawing DW, which is 32.992, is identified, and the z-direction dimension of the object in the drawing DW, which is 0.079, is identified.

[0034] The creation unit 22 processes the results obtained by the various trained models described above (for example, various data such as dimension values, dimension types, or recognized strings) into a more manageable format or performs editing to remove inappropriate data. For example, it removes the letter φ from the reading result "φ0.197" to obtain only the numerical data. For example, the creation unit 22 processes the results obtained by the various trained models into a more manageable format or performs editing to remove inappropriate data using a method such as the one disclosed in Japanese Patent Publication No. 7377565.

[0035] The creation unit 22 then creates related data as shown in Figure 3 by associating the various data described above. This related data is referenced in the search process described later. Note that the related data in Figure 3 is just one example. For example, a table with converted units may be stored in a separate table, or the converted dimension value information may not be stored and may be calculated each time a request is made. In addition, the creation unit 22 updates the related data by performing the above-described process on the image data each time new image data is received by the reception unit 20. As described above, the related data in this embodiment is data in which unit system data and numerical data obtained by recognizing at least the unit system and numerical value of the object drawn in each of the multiple image data are associated with the multiple image data. Furthermore, the unit system data and numerical data associated with the related data in this embodiment are data obtained by recognizing the unit system and numerical value of the image data by a trained model or the like for recognizing the unit system and numerical value of the object drawn in the image data.

[0036] The numerical data associated with the related data includes at least one of the dimensions, tolerances, surface roughness, weight, and volume values ​​of the object depicted in the image data. Furthermore, as mentioned above, the numerical data associated with the related data is data obtained in advance by recognizing the numerical values ​​depicted in the image data using a trained model or the like.

[0037] The search unit 24 searches for target image data that represents a drawing of an object and satisfies the conditions expressed by the unit system data, numerical data, and attribute data received by the reception unit 20. Specifically, it identifies target image data that satisfies the search conditions by searching multiple image data stored in the image database 30 based on the search conditions received by the reception unit 20.

[0038] Figures 5 and 6 show examples of screens displayed on the display unit (not shown) of the user terminal 18. The user searches for target image data that meets the search conditions they have set by operating the screens shown in Figures 5 and 6. For example, the user performs a search for target image data that meets the search conditions represented by the keyword by entering a keyword in the "Enter keyword" field of the input screen S1 in Figure 5. Figure 5 shows the search results R of target image data that meet the search conditions. In addition, for example, the user can display another input screen S2, as shown in Figure 6, on the display unit (not shown) of the user terminal 18 by operating the input screen S1.

[0039] As shown in Figure 6, the input screen S2 allows you to set search conditions related to dimension J1 as basic conditions. As shown in Figure 6, by checking the checkboxes for thickness, outer diameter, width, height, and depth of dimension J1, you can set search conditions related to the attributes of thickness, outer diameter, width, height, and depth. These checkboxes may be either one-choice or allow multiple selections. Furthermore, it may be possible to control the selection so that dimension values ​​that are generally incompatible in the same drawing, such as thickness and outer diameter, cannot be selected simultaneously. As shown in J3 in Figure 6, the user can set numerical data for dimensions by operating a slider bar. In addition, a field for direct input of numerical data may be provided, or both may be displayed simultaneously.

[0040] For example, as shown in Figure 6, if the user checks the checkboxes for the attributes width, height, and depth, the detailed conditions shown in Figure 6 will be displayed, and it will be possible to input numerical data related to width, height, and depth. Also, as shown in Figure 6, it is possible to specify the unit system to which the drawing conforms. In the example shown in Figure 6, it is possible to specify mm, inch, and feet as the unit system to which the drawing conforms. For example, as shown in J3 of Figure 6, if 45 to 55 is specified as the numerical data for the attributes width, height, and depth, and as shown in J4, mm is specified as the unit system to which the drawing conforms, then the image data of drawings in which the width, height, and depth of the object drawn in the drawing are between 45 mm and 55 mm will be searched from among the multiple image data stored in the image database 30. Specifically, after making the above input or selection, the user can click the "Refine" button in Figure 6 to execute the search. The user can also exit screen S2 at any time and return to screen S1 in Figure 5 by clicking the "Cancel" button in Figure 6. Although not shown in Figures 5 and 6, buttons such as download buttons may be placed on the screen to allow users to download dimensional information.

[0041] By operating screens as shown in Figures 5 and 6, users can search for image data or download dimensional information from the image database 30 using unified dimensional conditions. As explained above, the related data holds dimensional values ​​converted from the unit system to which the drawing conforms to to another unit system. Therefore, by simply specifying the search conditions as described above, users can uniformly search the image data in the image database 30 without any input burden.

[0042] When searching for target image data that satisfies a search condition from among a plurality of pieces of image data stored in an image database (30), the search unit (24) searches for target image data satisfying the above-described search condition from among the plurality of pieces of image data stored in the image database (30) by referring to related data stored in a related data storage unit (32). For this reason, the search unit (24) refers to the related data to search for target image data that satisfies the search condition from among the plurality of pieces of image data stored in the image database (30).

[0043] Note that if the unit system data accepted by the accepting unit (20) is different from the unit system that the image data in the image database (30) conforms to, the search unit (24) acquires, as a search result, target image data in which the numerical value obtained by converting the numerical value in the image data to the unit system represented by the accepted unit system data corresponds to the accepted numerical data. Specifically, when searching for target image data that satisfies the search condition, the search unit (24) searches for target image data satisfying the search condition from among the plurality of pieces of image data stored in the image database (30) by referring to not only the dimension type and dimension value of the related data but also the converted unit system and converted dimension value. For example, if numerical data representing a dimension value of a target object is specified as 1 inch to 2 inches as a search condition, a drawing in which the dimension value of the target object is 25.4 mm to 50.8 mm is also acquired as a search result.

[0044] For this reason, according to the information processing apparatus (16) of the present embodiment, it is possible to acquire image data in which a target object drawn in a drawing satisfies a predetermined dimension condition from among a plurality of pieces of image data in which different unit systems are mixed.

[0045] The output unit (26) outputs the target image data searched by the search unit (24) as a search result. Specifically, the output unit (26) outputs target image data that satisfies the condition represented by the unit system data and numerical data accepted by the accepting unit (20). For example, the output unit (26) outputs a search result R as shown in FIG. 5.

[0046] (User Terminal 18) User terminal 18 is a terminal operated by a user. Specifically, a user exchanges information with the information processing apparatus 16 by operating the user terminal 18.

[0047] Each of the information processing apparatus 16 and the user terminal 18 of the information processing system 10 can be implemented, for example, by a computer 70 shown in FIG. 7. The computer 70 includes a CPU 71, a memory 72 serving as a temporary storage area, and a non-volatile storage unit 73. The computer 70 also includes an input-output interface (I / F) 74 to which an input-output device and the like (not shown) are connected, and a read / write (R / W) unit 75 that controls reading and writing of data to and from a recording medium. The computer 70 further includes a network I / F 76 connected to a network such as the Internet. The CPU 71, the memory 72, the storage unit 73, the input-output I / F 74, the R / W unit 75, and the network I / F 76 are connected to each other via a bus 77.

[0048] The storage unit 73 can be implemented by a Hard Disk Drive (HDD), a solid state drive (SSD), a flash memory, or the like. A program for causing the computer 70 to function is stored in the storage unit 73 serving as a storage medium. The CPU 71 reads the program from the storage unit 73, develops the program into the memory 72, and sequentially executes processes included in the program.

[0049] <Operation of Information Processing System 10> Next, the operation of the information processing system 10 according to the present embodiment will be described.

[0050] (Related Data Creation Process) First, the information processing device 16 creates related data based on multiple image data. The user operates their user terminal 18 to send image data to the information processing device 16. The receiving unit 20 of the information processing device 16 receives the image data, assigns a drawing ID to the image data, and stores it in the image database 30. Each time the information processing device 16 receives new image data, it assigns a drawing ID to that image data and stores it in the image database 30. As a result, multiple image data files are stored in the image database 30.

[0051] When the information processing device 16 receives a signal instructing the creation of related data, it executes the related data creation processing routine shown in Figure 8.

[0052] In step S100, the creation unit 22 acquires multiple image data by reading multiple image data stored in the image database 30.

[0053] In step S102, the creation unit 22 identifies the unit system, dimension value, and dimension type of the image data by inputting each of the multiple image data acquired in step S100 into the trained model storage unit 28, which is stored in the trained model storage unit 28, a trained model for unit system discrimination, a trained model for dimension recognition, and a trained model for character image recognition.

[0054] In step S104, the creation unit 22 calculates a converted dimension value for each of the multiple image data by converting the dimension value identified in step S102 to a dimension value in a different unit system. For example, if the unit system identified in step S102 is inches, the dimension value conforming to inches identified in step S102 is converted to a dimension value conforming to millimeters, which is an example of another unit system.

[0055] In step S106, the creation unit 22 creates associated data as shown in Figure 3 by associating the unit system, dimension value, and dimension type identified in step S102 with the converted dimension value obtained in step S104 for each of the multiple image data.

[0056] In step S108, the creation unit 22 stores the related data obtained in step S106 into the related data storage unit 32 and terminates this processing routine.

[0057] In the routine shown in Figure 8, we have explained the case where related data is created from each of the multiple image data stored in the image database 30 as an example. However, it is also possible to update the existing related data each time new image data is received.

[0058] (Image data search processing) Once related data is created by the routine in Figure 8 and stored in the related data storage unit 32, it becomes possible to search for image data using the related data. When the information processing device 16 receives a signal instructing it to search for image data, it executes the search processing routine shown in Figure 9.

[0059] In step S200, the reception unit 20 receives the specification of unit system data representing a unit system, the specification of numerical data conforming to the unit system data, and the specification of attribute data representing the attributes of the numerical data. The reception unit 20 displays an input screen for receiving the specification of unit system data, numerical data, and attribute data on the display unit (not shown) of the user terminal 18, and receives the unit system data, numerical data, and attribute data entered by the user operating the user terminal 18. For example, the reception unit 20 displays an input screen S2 as shown in Figure 6 on the display unit (not shown) of the user terminal 18, and receives the unit system data specified by the user as shown in J4. Alternatively, for example, the reception unit 20 receives numerical data specified by the user that conforms to the unit system data, as shown in J3 in Figure 6.

[0060] In step S202, the search unit 24 searches the image database 30, which stores multiple image data, for target image data that represents a drawing of an object and satisfies the conditions represented by the unit system data, numerical data, and attribute data received in step S200.

[0061] In step S204, the output unit 26 outputs the target image data obtained in step S202 as a search result. Specifically, the output unit 26 displays an output screen showing the target image data on the display unit (not shown) of the user terminal 18. For example, the output unit 26 displays an output screen like the one shown in S1 of Figure 5 on the display unit (not shown) of the user terminal 18.

[0062] The search results output from the output unit 26 are transmitted to the user terminal 18. The user checks the search results displayed on the display unit (not shown) of their user terminal 18.

[0063] As described above, the information processing device according to this embodiment accepts the specification of unit system data representing a unit system, the specification of numerical data conforming to said unit system data, and the specification of attribute data. The information processing device outputs target image data that represents a drawing of an object and satisfies the conditions represented by the unit system data, numerical data, and attribute data. This makes it possible to obtain image data from image data created in accordance with different unit systems in which the depicted object satisfies predetermined numerical conditions.

[0064] Furthermore, at least some of the image data stored in the image database of this embodiment do not have the corresponding unit system indicated. In addition, if the received unit system data and the unit system to which the image data in the image database conforms are different, the information processing device of this embodiment retrieves the target image data as a search result, where the numerical value obtained by converting the numerical value in the image data in the image database to the unit system represented by the received unit system data corresponds to the received numerical data. This makes it possible to retrieve image data that satisfies predetermined numerical conditions even when multiple image data stored in the image database conform to different unit systems.

[0065] Furthermore, according to this embodiment, a single query can extract the data desired by the user, using unit system data such as millimeters or inches and numerical data (e.g., length) that conforms to that unit system as conditions. For example, by specifying the conditions "300" and "mm", it is possible to extract image data written in millimeters where the string "300" is written on the drawing, and image data written in inches where the string "11.811" is written on the drawing. Alternatively, by specifying the conditions "diameter", "from "45" to "55"", and "mm", it is possible to extract image data written in millimeters where the string corresponding to the diameter dimension on the drawing is in the range of 45-55, and image data written in inches where the string corresponding to the diameter dimension on the drawing is in the range of 1.772-2.165.

[0066] This disclosure is not limited to the embodiments described above, and various modifications and applications are possible without departing from the gist of this disclosure.

[0067] For example, in the above embodiment, the case in which the information processing device 16 itself comprises an image database 30 and a related data storage unit 32 was described as an example, but it is not limited to this. For example, the information processing device 16 may acquire various data from an external database or external storage unit different from the information processing device 16 via the network 19. Also, in the above embodiment, the case in which various information in image data is recognized using multiple trained models was described as an example, but it is not limited to this. Various information in image data may be recognized using a single trained model, or various information in image data may be recognized using only a rule-based method.

[0068] Furthermore, the output unit 26 of the above embodiment may be configured to highlight the area of ​​numerical values ​​included in the target image data, which is the search result. Specifically, although not shown in the search result R of Figure 5 above, it may be configured to highlight the location of numerical values ​​on the drawing of the search result R that correspond to the numerical data used for the search. As a method for identifying the highlighted location, one possible method is to store the coordinate information of the location when detecting the broad-sense dimension information mentioned above, and then highlight the area indicated by these coordinates. For example, if the search condition is specified as the plate thickness of the object drawn in the drawing being between 2 mm and 5 mm, then in a drawing conforming to millimeters, if the numerical value determined to correspond to the plate thickness is 2.0 or 2.5, then this location on the drawing will be highlighted. Similarly, in a drawing conforming to inches, if the numerical value determined to correspond to the plate thickness is 0.079, this corresponds to 2.0 mm, and will be highlighted in the same way. In this way, in this embodiment, even if each of the multiple drawings conforms to a different unit system, it is possible to search for image data that satisfies the search condition by using dimension type information that is not directly described in the drawing. Furthermore, users can search for image data uniformly without any special input burden, providing them with the convenience of finding the desired image data without the input burden of registering search criteria. The user interface using a slider bar also reduces the input burden during searches, allowing users to easily find image data. Additionally, the ability to download dimension information from drawings allows users to uniformly perform calculations, for example, on drawings with different unit systems, using spreadsheet software without additional input burden. Moreover, the highlighting of search results allows users to intuitively understand where the numerical data they searched for is located within the drawing, contributing to their work efficiency.

[0069] Furthermore, the image data representing the drawing of the object in the above embodiment may not be limited to image data obtained by converting a design drawing created using CAD into PDF format, but may also be image data created by other means. For example, it may be image data obtained by scanning a paper drawing.

[0070] Furthermore, in the above embodiment, each process that the CPU reads and executes software (programs) may be executed by various processors other than the CPU. Examples of such processors include PLDs (Programmable Logic Devices) whose circuit configuration can be changed after manufacturing, such as FPGAs (Field-Programmable Gate Arrays), and dedicated electrical circuits that have a circuit configuration specifically designed to execute a particular process, such as ASICs (Application Specific Integrated Circuits). Each process may be executed by one of these various processors, or by a combination of two or more processors of the same or different types (for example, multiple FPGAs, and a combination of a CPU and an FPGA). More specifically, the hardware structure of these various processors is an electrical circuit that combines circuit elements such as semiconductor elements.

[0071] Furthermore, although the above embodiment describes a configuration in which each program is pre-stored (installed) in a storage device, the invention is not limited to this. The program may be provided in a form stored on a storage medium such as a CD-ROM, DVD-ROM, Blu-ray disc, or USB memory. Alternatively, the program may be provided in a form that is downloaded from an external device via a network.

[0072] (Note) The following is a note regarding the nature of this disclosure.

[0073] (Note 1) An information processing device comprising: a receiving unit that accepts the specification of unit system data representing a unit system, the specification of numerical data conforming to the unit system data, and the specification of attribute data; and an output unit that outputs target image data representing a drawing of an object, which satisfies the conditions represented by the unit system data, the numerical data, and the attribute data. (Note 2) The information processing device according to Note 1, further comprising a search unit that searches for the target image data from a database in which image data representing a drawing of an object is stored, wherein the output unit outputs the target image data retrieved by the search unit. (Note 3) The information processing device according to Note 2, wherein when the search unit searches for the target image data from a plurality of the image data stored in the database, it searches for the target image data by referring to related data which associates the unit system data, the numerical data, the attribute data, and the plurality of the image data, obtained by recognizing at least the unit system, numerical value, and attribute of the object drawn in each of the plurality of the image data. (Note 4) The information processing apparatus according to Note 3, wherein the unit system data, numerical data, and attribute data associated with the related data are data obtained by recognizing the unit system, numerical data, and attributes of the image data by an image recognition means for recognizing the unit system and numerical data of an object drawn in the image data. (Note 5) The information processing apparatus according to Note 3, wherein the attribute data includes at least one of the dimensional value, tolerance, surface roughness, weight value, and volume value of an object drawn in the image data, and is data obtained in advance by recognizing each attribute of each description drawn in the image data by an image recognition means for recognizing each attribute of each description. (Note 6) The information processing apparatus according to Note 2, wherein if the unit system data received by the receiving unit and the unit system to which the image data in the database conforms are different, the search unit obtains the target image data as a search result in which the numerical value obtained when the numerical value in the image data is converted to the unit system represented by the received unit system data corresponds to the received numerical data.(Note 7) The information processing device according to Note 2, wherein at least some of the image data of the plurality of image data stored in the database does not have a corresponding unit system described. (Note 8) The information processing device according to Note 1 or Note 2, wherein the receiving unit displays an input screen on the user terminal for receiving the specification of the unit system data, the specification of the numerical data, and the specification of the attribute data, and receives the unit system data and the numerical data entered by the user operating the user terminal, and the output unit displays an output screen on the user terminal showing the target image data. (Note 9) The information processing device according to Note 1 or Note 2, wherein the data format of the image data is at least one of raster data format and vector data format. (Note 10) The information processing device according to Note 1 or Note 2, wherein the output unit highlights the area of ​​the numerical data included in the target image data. (Note 11) An information processing method in which a computer performs processing that accepts the specification of unit system data representing a unit system, the specification of numerical data conforming to the unit system data, and the specification of attribute data, and outputs target image data that represents a drawing of an object and satisfies the conditions expressed by the unit system data, the numerical data, and the attribute data. (Note 12) An information processing program for causing a computer to perform processing that accepts the specification of unit system data representing a unit system, the specification of numerical data conforming to the unit system data, and the specification of attribute data, and outputs target image data that represents a drawing of an object and satisfies the conditions expressed by the unit system data, the numerical data, and the attribute data.

[0074] All documents, patent applications, and technical standards described herein are incorporated by reference to the same extent as if each individual document, patent application, and technical standard were specifically and individually noted to be incorporated by reference.

Claims

1. An information processing device comprising: a receiving unit that accepts the specification of unit system data representing a unit system, the specification of numerical data conforming to the unit system data, and the specification of attribute data representing the attributes of the numerical data; and an output unit that outputs target image data representing a drawing of an object, which satisfies the conditions represented by the unit system data, the numerical data, and the attribute data.

2. The information processing apparatus according to claim 1, further comprising a search unit that searches for target image data from a database storing image data representing a drawing of an object, wherein the output unit outputs the target image data retrieved by the search unit.

3. The information processing apparatus according to claim 2, wherein the search unit searches for the target image data from among a plurality of image data stored in the database, by referring to related data which is obtained by recognizing at least the unit system, numerical value, and attributes of the object drawn on each of the plurality of image data, and which is associated with the unit system data, numerical value data, attribute data, and the plurality of image data.

4. The information processing apparatus according to claim 3, wherein the unit system data, numerical data, and attribute data associated with the related data are data obtained by recognizing the unit system, numerical data, and attributes of the image data by an image recognition means for recognizing the unit system, numerical data, and attributes of an object drawn in the image data.

5. The information processing apparatus according to claim 3, wherein the attribute data includes at least one of the dimensional value, tolerance, surface roughness, weight value, and volume value of the object depicted in the image data, and is data obtained in advance by an image recognition means for recognizing each attribute of the description depicted in the image data.

6. The information processing apparatus according to claim 2, wherein if the unit system data received by the receiving unit is different from the unit system to which the image data stored in the database conforms, the search unit obtains as a search result the target image data to which the converted numerical value obtained when the numerical value in the image data is converted to the unit system represented by the received unit system data corresponds to the received numerical data.

7. The information processing apparatus according to claim 2, wherein at least some of the image data stored in the database does not have a corresponding unit system described.

8. The information processing apparatus according to claim 1 or 2, wherein the receiving unit displays an input screen on the user terminal for receiving the specification of the unit system data, the specification of the numerical data, and the specification of the attribute data, and receives the unit system data and the numerical data input by the user operating the user terminal, and the output unit displays an output screen on the user terminal in which the target image data is displayed.

9. The information processing apparatus according to claim 2, wherein the data format of the image data is at least one of raster data format and vector data format.

10. The information processing apparatus according to claim 1 or claim 2, wherein the output unit highlights the area of ​​numerical values ​​included in the target image data.

11. An information processing method in which a computer performs processing that accepts the specification of unit system data representing a unit system, the specification of numerical data conforming to the unit system data, and the specification of attribute data representing the attributes of the numerical data, and outputs target image data representing a drawing of an object, which satisfies the conditions represented by the unit system data, the numerical data, and the attribute data.

12. An information processing program for causing a computer to perform a process that accepts the specification of unit system data representing a unit system, the specification of numerical data conforming to the unit system data, and the specification of attribute data representing the attributes of the numerical data, and outputs target image data representing a drawing of an object, which satisfies the conditions represented by the unit system data, the numerical data, and the attribute data.

Images