Machining information generation device, program, and machining information generation method

Abstract

A machining information generation device (100) comprises: a CAD data drawing unit (102) that uses design data indicating design details about an object to create a line drawing in which the exterior shape of the object is indicated with lines; a color fill region identification unit (103) that identifies a plurality of partial regions from the line drawing by defining, as a single partial region, a single region enclosed by lines in a portion of the line drawing where machining of the object is to be performed; a segmentation unit (105) that uses the design data to perform classification into a plurality of machining regions by performing semantic segmentation to classify a region where machining is to be performed in one step as a single machining region; and a fill processing unit (106) that applies a correction to the classification results such that each of the plurality of partial regions is classified into one of the plurality of machining regions.

Description

Processing information generation device, program, and processing information generation method

[0001] The present disclosure relates to a processing information generation device, a program, and a processing information generation method.

[0002] In NC (Numerical Control) machining, machining areas to be NC machined are identified from the external shape of the target object represented by CAD (Computer Aided Design) data, and a path is identified for each machining area.

[0003] When identifying a machining area from the external shape of an object to be NC machined, for example, semantic segmentation inference may be performed using an AI (Artificial Intelligence) model. Specifically, the machining area can be identified using the technology described in Non-Patent Document 1.

[0004] Olaf Ronneberger, Philipp Fischer, and Thomas Brox, “U-Net: Convolutional Networks for Biomedical Image Segmentation”, arXiv:1505.04597v1 [cs. CV], June 18, 2015

[0005] However, conventional semantic segmentation using AI models classifies each pixel by taking into account the shape to be processed and the surrounding conditions, etc. Therefore, differences in the shape to be processed and the conditions of the surrounding pixels may result in pixels being classified into different classes even though they are the same processing area.

[0006] Therefore, one or more aspects of the present disclosure aim to enable more accurate identification of a processing area from the external shape of an object.

[0007] A processing information generation device according to one aspect of the present disclosure is characterized by comprising: a drawing unit that uses design data indicating the design content of an object to be processed to draw a line drawing showing the external shape of the object with lines; a partial area identification unit that identifies multiple partial areas from the line drawing by designating each area surrounded by lines in the part of the line drawing where processing of the object is to be performed as one partial area; a segmentation unit that classifies multiple processing areas by using the design data to perform semantic segmentation to classify areas to be processed in one process as one processing area; and a correction unit that corrects the classification results into the multiple processing areas so that each of the multiple partial areas is classified into one of the multiple processing areas.

[0008] A program according to one aspect of the present disclosure causes a computer to function as a drawing unit that uses design data indicating the design content of an object to be processed to draw a line drawing showing the external shape of the object with lines; a partial area identification unit that identifies multiple partial areas from the line drawing by designating each area surrounded by lines in the part of the line drawing where the object is to be processed as one partial area; a segmentation unit that classifies multiple processing areas by using the design data to perform semantic segmentation to classify areas where processing is to be performed in one process as one processing area; and a correction unit that corrects the classification results into the multiple processing areas so that each of the multiple partial areas is classified into one of the multiple processing areas.

[0009] A processing information generation method according to one aspect of the present disclosure includes using design data indicating the design content of an object to be processed to draw a line drawing showing the external shape of the object with lines, and identifying multiple partial areas from the line drawing by treating each area surrounded by lines in the part of the line drawing where processing of the object is to be performed as one partial area, and using the design data to perform semantic segmentation to classify areas to be processed in one process into one processing area, thereby classifying the multiple processing areas, and correcting the classification results into the multiple processing areas so that each of the multiple partial areas is classified into one of the multiple processing areas.

[0010] According to one or more aspects of the present disclosure, it is possible to more accurately identify a processing area from the external shape of an object.

[0011] FIG. 1 is a block diagram schematically showing the configuration of a processing information generating device according to a first embodiment. FIG. 2 is a schematic diagram showing an example of a line drawing of a processing object. FIG. 3 is a schematic diagram showing an example of an area classification image in which the processing object is classified into processing areas. FIG. 4 is a first schematic diagram for explaining processing in a fill processing unit. FIG. 5 is a second schematic diagram for explaining processing in a fill processing unit. FIG. 6 is a block diagram schematically showing the configuration of a PC. FIG. 7 is a flowchart showing the operation of the processing information generating device according to the first embodiment. FIG. 8 is a block diagram schematically showing the configuration of a processing information generating device according to a second embodiment. FIG. 9 is a schematic diagram showing an example of a graph generated in a periphery correction unit. FIG. 10 is a schematic diagram for explaining processing in the periphery correction unit. FIG. 11 is a flowchart showing the operation of the processing information generating device according to the second embodiment. FIG. 12 is a block diagram schematically showing the configuration of a processing information generating device according to a third embodiment. FIG. 13 is a flowchart showing the operation of the processing information generating device according to the third embodiment.

[0012] 1 is a block diagram showing a schematic configuration of a processing information generation device 100 according to embodiment 1. The processing information generation device 100 includes a CAD data storage unit 101, a CAD data drawing unit 102, a coloring area identification unit 103, a trained model storage unit 104, a segmentation unit 105, and a filling processing unit 106.

[0013] The CAD data storage unit 101 stores CAD data as design data indicating the design content of a processing object to be processed. The CAD data indicates, for example, the dimensions and shape of the processing object as the design content.

[0014] The CAD data drawing unit 102 is a drawing unit that uses CAD data to draw a line drawing that shows the external shape of the object to be machined with lines. Here, the drawing unit draws a line drawing of the external shape of the object to be machined when viewed from one direction. For example, the CAD data drawing unit 102 references the CAD data and generates a line drawing of the object to be machined by drawing, using multiple lines, the external shape of the object to be machined from the direction of machining by NC machining.

[0015] 2 is a schematic diagram showing an example of a line drawing of an object to be processed. As shown in line drawing 120 of the object to be processed, the object to be processed is to be processed by cutting predetermined first shapes 122 and second shapes 123 into a base material 121.

[0016] Returning to Figure 1, the coloring area identification unit 103 is a partial area identification unit that identifies multiple coloring areas from the line drawing drawn by the CAD data drawing unit 102 by identifying one area surrounded by lines in the part of the object to be processed as a coloring area as one partial area.

[0017] For example, the coloring region identification unit 103 identifies coloring regions R01 to R16 from the line drawing 120 to be processed, as shown in Figure 2. As described above, the base material 121 portion of the processing target is not processed, and therefore this portion is not identified as a coloring region.

[0018] Returning to Figure 1, the trained model storage unit 104 stores trained models for performing semantic segmentation.

[0019] The segmentation unit 105 classifies multiple machining regions by performing semantic segmentation using the trained model stored in the trained model storage unit 104 and CAD data of the machining target, which classifies regions to be machined in one process into one machining region. The segmentation unit 105 here may perform known semantic segmentation. The segmentation unit 105 generates a region classification image by coloring each machining region with a different color.

[0020] Specifically, the segmentation unit 105 generates a depth map, which is a depth image indicating the depth at the time of processing, from the shape of the processing object indicated by the CAD data. Then, the segmentation unit 105 inputs the generated depth map to the trained model, thereby acquiring, from the trained model, a region classification image that is the result of classification into multiple processing regions.

[0021] For example, the segmentation unit 105 generates a depth map as a machining shape image showing machining shapes for machining a first shape 122 and a second shape 123 based on the depth from the surface 121a of the base material 121 in the line drawing 120 of the machining target shown in Fig. 2. Then, the segmentation unit 105 inputs the generated depth map into the trained model to obtain a region classification image as shown in Fig. 3. In the region classification image shown in Fig. 3, the first shape 122 is classified into two machining regions R20 and R21.

[0022] 1 , the fill processing unit 106 is a correction unit that corrects the classification results of the segmentation unit 105 so that each of the multiple partial regions is classified into one of the multiple processing regions. Here, the fill processing unit 106 corrects the processing regions classified by the segmentation unit 105 using the coloring regions identified by the coloring region identification unit 103.

[0023] For example, the fill processing unit 106 sequentially identifies one partial area from among multiple partial areas as a correction target partial area. Next, if the correction target partial area is classified into two or more processing areas, the fill processing unit 106 selects one processing area with the largest area among the two or more processing areas. Then, the fill processing unit 106 corrects the classification result so that the correction target partial area is classified into the selected processing area.

[0024] Specifically, as shown in Figure 4, of the color-painting areas R01 to R16 identified by the color-painting area identification unit 103, three color-painting areas R04, 05, and 06 are classified by the segmentation unit 105 into two processing areas R20 and R21.

[0025] Normally, a portion determined to be a single color-filled area based on the appearance of the CAD data should be classified as a single processing area. Therefore, when a color-filled area is classified into multiple processing areas, the fill processing unit 106 corrects the multiple processing areas into a single processing area. For example, the fill processing unit 106 classifies a single color-filled area into the processing area with the largest area, in other words, the largest number of dots. As a result, for example, as shown in Figure 5, color-filled areas R04, R05, and R06 are classified as processing area R20.

[0026] The fill processing unit 106 then generates a processing image showing the corrected processing areas in the respective colors. Here, one processing area shown in the processing image indicates a part to be processed in one pass in NC processing.

[0027] The processing information generating device 100 described above can be realized by, for example, a computer such as the PC 10 shown in Fig. 6. The PC 10 includes a storage 11 such as a hard disk drive (HDD) and a solid state drive (SSD), a memory 12, a processor 13 such as a central processing unit (CPU), a communication interface (I / F) 14 such as a network interface card (NIC), an input interface 15 such as a keyboard and a mouse, and a display 16.

[0028] For example, the CAD data storage unit 101 and the learned model storage unit 104 can be realized by the storage 11 or the memory 12. The CAD data drawing unit 102, the coloring area identification unit 103, the segmentation unit 105, and the filling processing unit 106 can be realized by the processor 13 executing a program.

[0029] The program may be downloaded to the storage 11 from a recording medium (not shown) via a reader / writer (not shown) or from a network via the communication I / F 14, and then loaded onto the memory 12 and executed by the processor 13. Alternatively, the program may be directly loaded onto the memory 12 from a recording medium via the reader / writer or from a network via the communication I / F 14, and then executed by the processor 13. In other words, the program may be provided by a computer program product such as a recording medium.

[0030] 7 is a flowchart showing the operation of the machining information generating device 100 according to embodiment 1. First, the CAD data drawing unit 102 refers to the CAD data stored in the CAD data storage unit 101, and draws the external shape of the object to be machined from the direction of machining by NC machining using a plurality of lines, thereby generating an external image showing a line drawing of the object to be machined (S10).

[0031] Next, the coloring area specifying unit 103 specifies an area surrounded by lines drawn by the CAD data drawing unit 102 as a partial area to be colored in the part of the processing object to be processed (S11).

[0032] The segmentation unit 105 also acquires a region classification image in which the image to be processed is divided into a plurality of processing regions by performing known semantic segmentation on the CAD data of the processing target using the trained model stored in the trained model storage unit 104 (S12). Here, the segmentation unit 105 generates the region classification image by coloring each processing region with a different color.

[0033] Next, the fill processing unit 106 corrects the processing area classified by the segmentation unit 105 using the coloring area identified by the coloring area identification unit 103 (S13). Here, if the coloring area is classified into multiple processing areas, the fill processing unit 106 corrects the multiple processing areas into a single area. Specifically, the fill processing unit 106 classifies one coloring area so that it becomes the processing area with the largest area, in other words, the largest number of dots, within the single coloring area.

[0034] As described above, according to the first embodiment, by correcting the inference results from semantic segmentation using the external shape indicated by the CAD data, it is possible to prevent parts that appear to have a single shape from being classified into multiple regions and processed using different paths.

[0035] 8 is a block diagram showing a schematic configuration of a processing information generation device 200 according to embodiment 2. The processing information generation device 200 includes a CAD data storage unit 101, a CAD data drawing unit 102, a coloring area identification unit 103, a trained model storage unit 104, a segmentation unit 105, a filling processing unit 106, and a periphery correction unit 207.

[0036] The CAD data storage unit 101, CAD data drawing unit 102, coloring area identification unit 103, learned model storage unit 104, segmentation unit 105, and filling processing unit 106 of the processing information generation device 200 according to embodiment 2 are similar to the CAD data storage unit 101, CAD data drawing unit 102, coloring area identification unit 103, learned model storage unit 104, segmentation unit 105, and filling processing unit 106 of the processing information generation device 100 according to embodiment 1. However, the filling processing unit 106 of the processing information generation device 200 according to embodiment 2 provides a corrected processing area image indicating the corrected processing area and a coloring area image indicating the coloring area identified by the coloring area identification unit 103 to the periphery correction unit 207.

[0037] The surrounding correction unit 207 sequentially selects a correction target partial area, which is one partial area to be corrected, from multiple partial areas, and after correction is made by the fill processing unit 106 in accordance with the processing area into which two or more partial areas surrounding the correction target partial area are classified, performs a correction to change the one processing area into which the correction target partial area is classified into another processing area.

[0038] For example, the periphery correction unit 207 identifies the color, which is the classification of each color-filled area shown in the color-filled area image, in the processing area shown in the corrected processing area image, and corrects the classification of the color-filled area using the classification of the surrounding color-filled areas. Specifically, after the correction by the fill processing unit 106, the periphery correction unit 207 changes the processing area into which the correction target partial area is classified to one processing area into which two or more partial areas surrounding the correction target partial area are classified most often.

[0039] Unlike the above, the surrounding correction unit 207 may change the one processing area into which the partial area to be corrected is classified to the one processing area with the largest area classified among two or more partial areas surrounding the partial area to be corrected, before correction is performed by the fill processing unit 106.

[0040] Here, the two or more partial areas are considered to be adjacent to the partial area to be corrected, but the two or more partial areas may also consist of an adjacent partial area that is adjacent to the partial area to be corrected, and a partial area adjacent to the adjacent partial area.

[0041] Specifically, the perimeter correction unit 207 regards the colored areas shown in the colored area image as nodes and determines whether the colored areas adjacent to each corresponding node are adjacent by an edge or a point. The perimeter correction unit 207 determines whether the adjacent nodes are adjacent by an edge as an edge junction and whether the adjacent nodes are adjacent by a point as a point junction, and determines the edge junctions and point junctions of all nodes. Then, the perimeter correction unit 207 generates a graph showing the adjacency relationships of all nodes by regarding the adjacency relationships of all nodes (here, the edge junctions and point junctions) as edges.

[0042] 9 is a schematic diagram showing an example of a graph showing the adjacency relationships of all nodes. In the graph 130 shown in FIG. 9, the nodes N corresponding to the regions R01 to R09 shown in FIG. Ｒ０１ ~N Ｒ０９ The subscripts of the capital N indicate the corresponding regions. In addition, in the graph 130, edge tangents are shown with solid lines and point tangents are shown with dashed lines.

[0043] Next, the surrounding correction unit 207 identifies one target node from all nodes one by one, and corrects the color of the target node using the classification of all nodes adjacent to the target node, in other words, the most commonly assigned color.

[0044] In the example shown in FIG. 9, the node N Ｒ０７ is classified as the processing region R21. However, the node N Ｒ０７ is a node N classified into the processing region R20. Ｒ０４ and node N Ｒ０５ Node N adjacent to and classified into the processing region R21 Ｒ０８ Node N is adjacent to Ｒ０７ Since there are more nodes classified as processing region R20 than nodes adjacent to node N Ｒ０７ is corrected to the processing region R20. Ｒ０８ and node N Ｒ０９ is also corrected to the processing region R20.

[0045] As described above, the correction by the periphery correction unit 207 corrects the corrected processing area image shown in FIG. 5 to the periphery corrected processing area image shown in FIG.

[0046] The processing information generating device 200 described above can also be realized by a computer such as the PC 10 shown in Fig. 6. For example, the periphery correction unit 207 can also be realized by the processor 13 executing a program.

[0047] Fig. 11 is a flowchart showing the operation of the processing information generating device 200 according to embodiment 2. Among the steps in the flowchart shown in Fig. 11, steps that perform the same processing as the steps in the flowchart shown in Fig. 7 are assigned the same reference numerals as in Fig. 7.

[0048] The processing of steps S10 to S13 in Fig. 11 is the same as the processing of steps S10 to S13 in Fig. 7. However, in Fig. 11, after step S13, the processing proceeds to step S24.

[0049] In step S24, the surrounding correction unit 207 identifies the color, which is the classification of each color-painted area shown in the color-painted area image, in the processing area shown in the corrected processing area image, and corrects the classification of the color-painted area using the classification of the surrounding color-painted areas.

[0050] As described above, according to the second embodiment, the inference results from semantic segmentation are corrected using the external shape shown in the CAD data, and further modified using the inference results from the surrounding areas, thereby enabling more accurate classification.

[0051] 12 is a block diagram showing a schematic configuration of a processing information generation device 300 according to embodiment 3. The processing information generation device 300 includes a CAD data storage unit 101, a CAD data drawing unit 102, a coloring area identification unit 103, a trained model storage unit 104, a segmentation unit 105, a filling processing unit 106, a user correction unit 308, a display unit 309, and an input unit 310.

[0052] The CAD data storage unit 101, CAD data drawing unit 102, coloring area specification unit 103, learned model storage unit 104, segmentation unit 105, and filling processing unit 106 of the processing information generation device 300 according to embodiment 3 are similar to the CAD data storage unit 101, CAD data drawing unit 102, coloring area specification unit 103, learned model storage unit 104, segmentation unit 105, and filling processing unit 106 of the processing information generation device 100 according to embodiment 1. However, the filling processing unit 106 of the processing information generation device 300 according to embodiment 3 provides a corrected processing area image indicating the corrected processing area to a user correction unit 308.

[0053] After the correction by the fill processing unit 106, the user correction unit 308 performs correction in accordance with a user instruction to change one processing area into which the partial area specified by the user is classified into another processing area. For example, the user correction unit 308 may display the multiple processing areas and the multiple partial areas after the correction by the fill processing unit 106 on the display unit 309, and may receive input of a correction instruction from the user via the input unit 310. Note that the display unit 309 may also display the multiple processing areas before such correction is performed.

[0054] Specifically, the user correction unit 308 displays the corrected processing area image from the fill processing unit 106 on the display unit 309, and accepts input of an instruction to correct the classification from the user via the input unit 310. Then, the user correction unit 308 generates a user-corrected processing area image that reflects the user's correction to the corrected processing area image.

[0055] The display unit 309 displays various images. In this example, the display unit 309 displays a correction processing area image. The input unit 310 accepts input of instructions from a user. In this example, the input unit 310 accepts input of correction instructions for the correction processing area image.

[0056] The processing information generating device 300 described above can also be realized by a computer such as the PC 10 shown in Fig. 6. For example, the user modification unit 308 can also be realized by the processor 13 executing a program. The display unit 309 can be realized by the display 16. The input unit 310 can be realized by the input I / F 15.

[0057] Fig. 13 is a flowchart showing the operation of the processing information generating device 300 according to embodiment 3. Among the steps in the flowchart shown in Fig. 13, steps that perform the same processing as the steps in the flowchart shown in Fig. 7 are assigned the same reference numerals as in Fig. 7.

[0058] The processing of steps S10 to S13 in Fig. 13 is the same as the processing of steps S10 to S13 in Fig. 7. However, in Fig. 13, after step S13, the processing proceeds to step S34.

[0059] In step S34, the user correction unit 308 causes the display unit 309 to display the correction processing area image.

[0060] Then, the user correction unit 308 determines whether or not the user has input a correction to the correction processing area image into the input unit 310 (S35). If the user has input a correction into the input unit 310 (Yes in S35), the process proceeds to step S36, and if the user has not input a correction into the input unit 310 (No in S35), the process proceeds to step S37.

[0061] In step S36, the user correction unit 308 corrects the image of the correction processing area in accordance with the correction instruction from the user, and the process then proceeds to step S37.

[0062] In step S37, the user correction unit 308 determines whether or not the user has input an end command into the input unit 310. If the user has input an end command into the input unit 310 (Yes in S37), the process ends, and if the user has not input an end command into the input unit 310 (No in S37), the process returns to step S35.

[0063] As described above, according to the third embodiment, the inference results from semantic segmentation are corrected using the external shape indicated by the CAD data, and further corrections are made in accordance with instructions from the user, thereby enabling more accurate classification.

[0064] In the third embodiment, the user's modifications are reflected in the corrected processing area image, which is the correction result by the fill processing unit 106 described in the first embodiment. However, the third embodiment is not limited to this example. For example, the user modification unit 308 may reflect the user's modifications in the surrounding modified processing area image modified by the surrounding modification unit 207 described in the second embodiment. In this case, the user modification unit 308, the display unit 309, and the input unit 310 may be added to the processing information generating device 200 shown in FIG. 8, and the user modification unit 308 may receive the surrounding modified processing area image from the surrounding modification unit 207. In other words, after the surrounding modification unit 207 has performed the modification, the user modification unit 308 may modify one processing area into which the partial area specified by the user is classified, to another processing area, in accordance with a user's instruction.

[0065] Furthermore, in the first to third embodiments described above, the segmentation unit 105 performs semantic segmentation by inputting a depth map to a trained model, but the first to third embodiments are not limited to such examples. The segmentation unit 105 can change the input data as appropriate depending on the configuration of the trained model. For example, CAD data may be the input data, or line drawing data representing a line drawing drawn by a CAD data drawing unit may be the input data.

[0066] 100, 200, 300 Processing information generation device, 101 CAD data storage unit, 102 CAD data drawing unit, 103 Coloring area identification unit, 104 Learned model storage unit, 105 Segmentation unit, 106 Filling processing unit, 207 Surrounding correction unit, 308 User correction unit, 309 Display unit, 310 Input unit.

Claims

1. A processing information generating device comprising: a drawing unit that uses design data indicating the design content of an object to be processed to draw a line drawing showing the external shape of the object with lines; a partial area identification unit that identifies multiple partial areas from the line drawing by treating each area surrounded by lines in the part of the line drawing where processing of the object will be performed as one partial area; a segmentation unit that uses the design data to perform semantic segmentation to classify areas to be processed in one process into multiple processing areas; and a correction unit that corrects the results of the classification into the multiple processing areas so that each of the multiple partial areas is classified into one of the multiple processing areas.

2. The processing information generation device described in claim 1, characterized in that the correction unit sequentially identifies one partial area from the multiple partial areas as a partial area to be corrected, and if the partial area to be corrected is classified into two or more processing areas, selects one processing area from the two or more processing areas that has the largest area among the partial areas to be corrected, and performs the correction so that the partial area to be corrected is classified into the selected processing area.

3. The processing information generating device according to claim 1 or 2, further comprising a surrounding correction unit that sequentially identifies one partial area from the plurality of partial areas as a partial area to be corrected, and, after the correction is performed, performs a correction to change the one processing area into which the partial area to be corrected is classified into another processing area according to the processing area into which two or more partial areas surrounding the partial area to be corrected are classified.

4. The processing information generation device described in claim 3, characterized in that the surrounding correction unit changes the one processing area into which the correction target partial area is classified to the one processing area into which the two or more partial areas are most frequently classified after the correction is performed.

5. The processing information generation device described in claim 3, characterized in that the surrounding correction unit changes the one processing area into which the partial area to be corrected is classified to the one processing area with the largest area classified among the two or more partial areas before the correction is performed.

6. The processing information generating device according to claim 4 or 5, wherein the two or more partial areas are adjacent to the partial area to be corrected.

7. A processing information generation device as described in claim 4 or 5, characterized in that the two or more partial areas consist of an adjacent partial area which is a partial area adjacent to the partial area to be corrected, and a partial area adjacent to the adjacent partial area.

8. The processing information generating device according to claim 1 or 2, further comprising a user correction unit that performs a correction after the correction is performed, in accordance with instructions from the user, to change one processing area into which the partial area specified by the user is classified, to another processing area.

9. The processing information generating device according to claim 8, further comprising: a display unit that displays the plurality of processing areas and the plurality of partial areas after the correction has been performed; and an input unit that accepts input of the instruction from the user.

10. The processing information generating device described in claim 8, further comprising: a display unit that displays the multiple processing areas after the correction has been performed, the multiple processing areas before the correction has been performed, and the multiple partial areas; and an input unit that accepts input of the instructions from the user.

11. The processing information generating device described in claim 3, further comprising a user correction unit that performs a correction after the correction has been made, in accordance with instructions from the user, to change one processing area into which the partial area specified by the user is classified, to another processing area.

12. A program that causes a computer to function as: a drawing unit that uses design data indicating the design content of an object to be processed to draw a line drawing showing the external shape of the object with lines; a partial area identification unit that identifies multiple partial areas from the line drawing by treating each area surrounded by lines in the part of the line drawing where processing of the object will be performed as one partial area; a segmentation unit that uses the design data to perform semantic segmentation to classify an area to be processed in one process into multiple processing areas; and a correction unit that corrects the results of the classification into the multiple processing areas so that each of the multiple partial areas is classified into one of the multiple processing areas.

13. A processing information generation method comprising the steps of: using design data indicating the design content of an object to be processed, drawing a line drawing showing the external shape of the object with lines; identifying multiple partial areas from the line drawing by defining each area surrounded by lines in the part of the line drawing where processing of the object will be performed as one partial area; using the design data, performing semantic segmentation to classify areas to be processed in one process into one processing area, thereby classifying the multiple processing areas; and correcting the classification results into the multiple processing areas so that each of the multiple partial areas is classified into one of the multiple processing areas.