Mold verification method

Abstract

This technology prevents the incorrect mold from being installed in the mold clamping device and also provides a mechanism to manage the mold and molding conditions in a linked manner. [Solution] As shown in Figure 1(a), an imaging mechanism 50 that photographs the top surface or a specific surface of the mold 40 and an image analysis unit 52 that analyzes the image taken by the imaging mechanism 50 are prepared, and the image analysis unit 52 determines the visual characteristics of the mold 40 by image recognition and verifies whether or not the mold 40 fixed to the mold clamping device 30 should be used for production.

Description

Technical Field

[0001] The present invention relates to a method for verifying a mold to verify whether the mold to be used is the correct mold.

Background Art

[0002] A variety of resin (plastic) molded products (hereinafter abbreviated as molded products) are manufactured by an injection molding apparatus. The injection molding apparatus mainly includes a mold clamping apparatus for clamping a mold, an injection apparatus for injecting a molten resin material into the mold, and a control unit for controlling these.

[0003] A mold is prepared for each molded product, set in the mold clamping apparatus, and removed from the mold clamping apparatus when the required number of molded products is obtained. As the types of molded products increase, the number and types of molds increase. The management of the number and types of molds becomes complicated. If an operator manages the molds based on a paper ledger or an electronic ledger, the cost related to management increases.

[0004] Therefore, various technologies for automating the management of molds in an injection molding apparatus have been put into practical use (for example, refer to Patent Document 1 (FIG. 1)).

[0005] In the injection molding system disclosed in Patent Document 1, first, an identification code is attached to the mold. Next, the identification code is read by a code reader. Next, it is confirmed whether the read identification code matches the identification code stored in the storage unit.

[0006] This reading operation is preferably performed in a state where the mold is fixed to the mold clamping apparatus. However, the mold clamping apparatus gets in the way, and it is not uncommon that the identification code is difficult to find or difficult to see. Therefore, the reading operation is often performed on the mold before it is attached to the mold clamping apparatus. This is because the reading operation becomes easier since it is not obstructed by the mold clamping apparatus.

[0007] Furthermore, multiple molds are often lined up according to the production plan in areas not obstructed by the mold clamping device. During the reading operation, the identification code of one of the multiple molds is read using a code reader. In other words, the location where the reading operation is performed is different from the location of the mold clamping device.

[0008] This can lead to human error, such as attaching a different mold to the clamping device than the one read by the code reader. The probability of this type of human error increases when the worker is unfamiliar with the procedure. Incorrect mold installation can trigger the following problems.

[0009] If the molding conditions (control conditions of the injection molding machine) are not set appropriately for the mold, it will be impossible to produce molded products. Furthermore, this can cause equipment malfunction. Incorrect mold installation can lead to improper setting of molding conditions, potentially resulting in such problems. Therefore, technology is needed to manage molds and molding conditions in a linked manner. [Prior art documents] [Patent Documents]

[0010] [Patent Document 1] Patent No. 7174134 [Overview of the Initiative] [Problems that the invention aims to solve]

[0011] The present invention aims to provide a technology that prevents the incorrect mold from being attached to the mold clamping device and that links and manages the mold and molding conditions. [Means for solving the problem]

[0012] The invention according to claim 1 includes an imaging mechanism for photographing the top surface or a specific surface of a mold, and an image analysis unit for analyzing the image captured by the imaging mechanism. The aforementioned image analysis unit identifies the visual characteristics of the mold by image recognition and verifies whether or not the mold fixed to the mold clamping device should be used for production. [Effects of the Invention]

[0013] In the invention according to claim 1, while the mold is fixed to the mold clamping device, it is checked whether or not the mold should be used for production. If, due to human error, the wrong mold is fixed to the mold clamping device, the image analysis unit detects the error. This detection prevents the wrong mold from being attached to the mold clamping device. When the correct mold is fixed to the mold clamping device, it becomes possible to link this mold with the molding conditions corresponding to this mold. Therefore, according to the present invention, a technology is provided that prevents the incorrect mold from being attached to the mold clamping device and also manages the mold and molding conditions in a linked manner. [Brief explanation of the drawing]

[0014] [Figure 1] (a) is a schematic diagram of the injection molding machine, (b) is a cross-sectional view of the mold, and (c) is a diagram showing the operation of the ejector pin. [Figure 2] (a) is an illustrative image of an image obtained by the imaging mechanism, (b) is an illustrative image of the same image as (a) but with increased magnification, and (c) is an illustrative image of the same image as (b) but with increased magnification. [Figure 3] (a) to (d) are diagrams illustrating the edges. [Figure 4] This is a master flow diagram according to the present invention. [Figure 5] This flowchart explains the details of the feature extraction process in the master flow. [Figure 6] This flowchart details the steps involved in extracting features from an image within the feature extraction process. [Figure 7] This is an illustrative diagram of a ledger. [Figure 8] This flowchart explains the details of the mold matching process in the master flow. [Embodiment for Carrying Out the Invention]

[0015] Embodiments of the present invention will be described below based on the accompanying drawings. [Examples]

[0016] [Injection Molding Device] As shown in Fig. 1(a), the injection molding device 10 includes a horizontally extending base 11, an injection device 20 supported by the base 11 so as to be movable in the horizontal direction, a mold clamping device 30 for clamping the mold 40 placed on the base 11, an imaging mechanism 50 for photographing the top surface or a specific surface of the mold 40, and an image analysis unit 52 for analyzing the image obtained by the imaging mechanism 50, and is a horizontal injection molding machine.

[0017] [Specific Surface] In the mold clamping device 30 in Fig. 1(a), the mold clamping shaft is horizontal. In this case, the imaging mechanism 50 photographs the "top surface" of the mold 40. The top surface is the same as the upper surface. In the case of a vertical injection molding machine, the mold clamping shaft is vertical. In this case, the imaging mechanism 50 photographs the "specific surface" of the mold 40. The specific surface is a surface different from the top surface of the mold and is a surface suitable for photographing, for example, the side surface or the front surface of the mold.

[0018] Also, the specific surface is, for example, a surface provided with bolt threading holes (reference numerals 55, 58, 59 in Fig. 2(a)) for screwing in suspension bolts. Also, the specific surface is not limited to one surface and may be a plurality of surfaces.

[0019] [Injection Device] For the sake of drawing convenience, only the heating cylinder 21 of the injection device 20 is shown. The resin material is kneaded and melted in the heating cylinder 21. The melted resin material is injected from the nozzle 22 at the tip into the mold 40.

[0020] [Mold Clamping Device] The clamping device 30 consists of a fixed platen 31 fixed to the base 11, a pressure receiving platen 32 fixed to the base 11, tie bars 33 connecting them, a movable platen 35 positioned between the pressure receiving platen 32 and the fixed platen 31 and resting on the base 11 via rails 34, and a clamping mechanism 36 that pushes the movable platen 35 closer to the fixed platen 31, using the pressure receiving platen 32 as a pivot point. The clamping mechanism 36 is preferably a hydraulic cylinder or a toggle link mechanism.

[0021] [Mold] The mold 40 consists of a fixed mold 41, a fixed mold mounting plate 42 for attaching the fixed mold 41 to a fixed platen 31, a movable mold 43, a movable mold mounting plate 44 for attaching the movable mold 43 to a movable platen 35, and an ejector mechanism 45 positioned between the movable mold mounting plate 44 and the movable mold 43.

[0022] [Transporting molds] The mold 40, having this structure, is suspended in its set state by suspension bolts and an electric winch, moved from the mold storage area to the mold clamping device 30, and then moved from the mold clamping device 30 back to the mold storage area.

[0023] [Uses of molds] Generally, molds known include press molds, casting molds, and resin molding molds. Furthermore, molds for resin molding include injection molding molds, extrusion molding molds, blow molding molds, vacuum molding molds, and compression molding molds. The mold 40 according to the present invention does not include a press mold. The mold 40 of the present invention includes casting molds (including die-casting molds) and resin molding molds.

[0024] As shown in Figure 1(b), a cavity 46 is provided between the fixed mold 41 and the movable mold 43. In the clamped state, molten resin material is injected into the cavity 46. At this time, the mold clamping device (Figure 1(a), reference numeral 30) is used to clamp the movable mold 43 so that it does not separate from the fixed mold 41 due to the injection pressure.

[0025] Next, as shown in Figure 1(c), the movable mold 43 is opened while the fixed mold 41 remains stationary. The ejector pin 47 of the ejector mechanism 45 is extended, releasing the molded product 48 from the movable mold 43. After this, the molded product 48 falls due to its own weight.

[0026] [Imaging mechanism] The imaging mechanism 50 shown in Figure 1(a) is a device that photographs the top surface or a specific surface of the mold 40, and is, for example, a CCD (charge-coupled device) camera. The optical axis (central axis) of the imaging mechanism 50 may be oblique. The method and location of installation of the imaging mechanism 50 are arbitrary, but for example, it can be attached to the fixed plate 31 via a suitable bracket.

[0027] [Image] The image quality obtained by capturing images with the imaging mechanism 50 will be explained based on Figures 2(a) to (c). As shown in Figure 2(a), Image 53 shows, for example, parts of the fixed mounting plate 42, the fixed type 41, the movable type 43, and the ejector mechanism 45.

[0028] A partition line 54, called the mold-fitting surface, is visible at the boundary between the fixed type 41 and the movable type 43. In the area of ​​the fixed type 41, a bolt threading hole 55 for screwing in a suspension bolt and two grooves 56 and 57 on the left and right are visible. In the movable area 43, two bolt threading holes 58 and 59 are visible. In the area of ​​the ejector mechanism 45, the notch 61 and the ejector pin 47 are visible.

[0029] The positions of bolt threading holes 55, 58, and 59 differ for each mold 40. The position and size of the notch 61 and ejector pin 47 also differ for each mold 40. The thickness of the fixed mold 41 and the thickness of the movable mold 43 also differ for each mold 40.

[0030] Figure 2(b) is a magnified view of a portion of Figure 2(a). As shown in Figure 2(b), by increasing the magnification, the previously inconspicuous streaks 56 and 57 became clearly visible, and a stain 62 appeared between the two streaks 56 and 57. In addition, several scratches 63 became visible.

[0031] Figure 2(c) is a magnified view of a portion of Figure 2(b). As shown in Figure 2(c), by further increasing the magnification, scratches 63 that were previously not very noticeable became clearly visible. In addition, cutting marks 64 that were previously not very noticeable became visible. Cutting marks 64 are also called tool marks or cutter marks.

[0032] The cutting marks 64 are arc-shaped marks that inevitably remain when the outer surface (including the top surface) of the fixed mold 41 is cut with a milling cutter or end mill. The milling cutter or end mill is supported by the spindle of the machine tool, but the spindle bends slightly due to the cutting reaction force. When the bending reaches a certain value, the spindle returns to its original position so that the bending becomes 0 (zero). Because bending and returning are repeated, the cutting marks 64 take on a wave-like pattern.

[0033] [Feature extraction] In this invention, cutting marks, mold outlines (shapes), and other features (feature points) are extracted using image processing. The method for extracting these features is arbitrary and not particularly limited. However, for the sole purpose of facilitating understanding, an example of image processing will be explained based on Figures 3(a) to (d).

[0034] Figure 3(a) shows the partition line 54. Macroscopically, the partition line 54 appears as a simple straight line, but microscopically, it is formed by a collection of points. Adjacent points have different brightness levels.

[0035] As shown in Figure 3(b), points where the brightness changes abruptly are recognized as edges 65. This recognition is performed by the image analysis unit (Figure 1, reference numeral 52). At the same time, the position coordinates of edges 65 are determined.

[0036] Figure 3(c) shows the cutting marks 64. Macroscopically, the cutting marks 64 appear as simple curves, but microscopically, they are formed by a collection of points. Adjacent points have different brightness levels. As shown in Figure 3(d), an edge 65 is also defined in the cutting mark 64. Such edges 65 make it possible to extract cutting marks 64 and the outline of the mold as features, or visual features.

[0037] Similarly, edges appear in the bolt threading holes 55, 58, and 59, the notch 61, the ejector pin 47, the contour lines of the fixed type 41, and the contour lines of the movable type 43, as shown in Figure 2(a). Furthermore, edges appear in the streaks 56, 57 and stains 62 shown in Figure 2(b).

[0038] Furthermore, an edge appears in the scratch 63 shown in Figure 3(c). In other words, an edge 65 is defined for any straight line, curve, or mixed straight line and curve line that appears in Figures 2(a) to (c).

[0039] Furthermore, the feature extraction method in this invention is not limited to the image processing described in Figures 3(a) to 3(d).

[0040] The image analysis unit 52 performs the following flow on an image obtained by the imaging mechanism 50 shown in Figure 1 (for example, Figure 2(a), reference numeral 53).

[0041] [Master Flow] As shown in Figure 4, the feature extraction process is performed at ST (indicating the step number; the same applies hereafter) 10. Next, the mold verification process is performed in ST20. The details of the feature extraction process in ST10 are explained in Figure 5 below.

[0042] [Details of the feature extraction process] In Figure 5, ST11, the mold is fixed to the mold clamping device. Incidentally, the temperature, pressure, and time in injection molding are naturally determined by the size and material of the molded product (Figure 1(c), reference numeral 48). Temperature, pressure, and time are called molding conditions. Therefore, in ST12, the molding conditions corresponding to this mold are input.

[0043] In addition, an ID (registration number, etc.) is assigned to this mold (ST13). In ST14, the assigned ID is linked to the molding conditions. ST15 is used to photograph the top surface or a specific surface of the mold. ST16 extracts features from the obtained images. The detailed flow of ST16 is explained based on Figure 6.

[0044] [Details of Feature Extraction] In ST151 in Figure 6, for example, features are extracted as shown in Figure 2(a). The extracted features are registered as the first features (ST152). Next, in ST153, for example, the image magnification is increased as shown in Figure 2(b). Then, in ST154, for example, features are extracted as shown in Figure 2(b). The extracted features are registered as second features (ST155).

[0045] Next, in ST156, the image magnification is further increased, for example, as shown in Figure 2(c). Then, in ST157, for example, features are extracted as shown in Figure 2(c). The extracted features are registered as third features (ST158).

[0046] [Number of feature extraction steps] In Figure 6, features were extracted in three steps: ST151, ST154, and ST157. The features differed at each step, and the quality and quantity of features increased with each of the three steps. As a result, the reliability of the matching in the subsequent mold matching process (Figure 4, ST20) is improved, which is desirable. However, while reliability increases, the processing in the image analysis unit (Figure 1, reference numeral 52) becomes more complex.

[0047] Therefore, although reliability is slightly reduced, extracting features in two steps is beneficial in order to reduce the burden on the image analysis unit (Figure 1, reference numeral 52). Furthermore, although reliability is further reduced, extracting features in a single step is also beneficial in order to further reduce the burden on the image analysis unit (Figure 1, reference numeral 52).

[0048] [When feature extraction is performed in one step] If feature extraction is performed in a single step, steps ST153 to ST158 are omitted in Figure 6. In other words, the step of extracting features from the image consists of ST151 and ST152. As a result, the number of steps in Figure 6 is drastically reduced, significantly lessening the burden on the image analysis unit (Figure 1, reference numeral 52).

[0049] [When feature extraction is performed in two steps] If feature extraction is performed in two steps, steps ST156 to ST158 are omitted in Figure 6. In other words, the steps for extracting features from the image consist of ST151 to ST155. As a result, the number of steps shown in Figure 6 is reduced, and the burden on the image analysis unit (Figure 1, reference numeral 52) can be lessened.

[0050] In ST17 of Figure 5, the obtained features (e.g., the first to third features) are registered along with an ID. An example of a ledger will be explained based on Figure 7.

[0051] [ledger] As shown in Figure 7, the ledger 67 contains a list of information such as the ID, molding conditions, and storage locations for the first to third features. Such a ledger 67 is an electronic ledger, which is stored in the image analysis unit (Figure 1(a), reference numeral 52) and displayed on an attached display.

[0052] With the above steps completed, the feature extraction process in ST10 shown in Figure 4 is finished. Next, the mold matching process in ST20 is performed. Details of this mold matching process are shown in Figure 8.

[0053] [Details of the mold matching process] In the manufacturing of molded products, the mold is determined according to the size and shape of the product, and the molding conditions corresponding to this mold are also determined. Therefore, in ST21 in Figure 8, the molding conditions B corresponding to mold A are input to the image analysis unit (Figure 1(a), reference numeral 52). In ST22, mold A is fixed to the clamping device. At this time, due to human error, the wrong mold may be fixed to the clamping device.

[0054] ST23 is used to photograph the top surface or a specific surface of mold A. In ST24, feature D is extracted from the obtained image C. This extraction is performed in accordance with the extraction in the previous feature extraction process (Figure 5, ST16). That is, if the extraction step in the previous feature extraction process is as shown in Figure 6, the extraction in ST24 is also performed according to Figure 6. As a result, feature D is composed of, for example, the first feature D1, the second feature D2, and the third feature D3.

[0055] In ST25, IDs matching feature D (first to third features D1 to D3) are identified from the ledger (Figure 7). Once the IDs are identified, the molding conditions corresponding to the identified IDs are determined from the ledger (Figure 7). Therefore, ST26 checks whether the molding conditions corresponding to the identified ID match the molding conditions B (ST21) of mold A.

[0056] If the result in ST26 is YES (match), then mold A, which is fixed to the mold clamping device, has been verified and confirmed to be the correct mold to be used for production (e.g., injection molding), and the process proceeds to ST27, where the automatic operation start interlock is released. From there, a predetermined number of molded parts are produced using mold A.

[0057] If the answer in ST26 is NO (not OK), then mold A, which is fixed to the clamping device, is determined to be a mold that should not be used for production (e.g., injection molding), i.e., an incorrect mold, and an alarm is issued (ST28). Another mold is prepared (ST29), and the authenticity of this other mold is verified using steps ST22 to ST26.

[0058] Based on the above description, the present invention can be summarized as follows. As shown in Figure 1(a), an imaging mechanism 50 that photographs the top surface or a specific surface of the mold 40 and an image analysis unit 52 that analyzes the image captured by the imaging mechanism 50 are prepared. The image analysis unit 52 identifies the visual characteristics of the mold 40 (Figures 3(b), (d), reference numeral 65) by image recognition, and a mold verification method is provided which verifies whether or not the mold 40 fixed to the mold clamping device 30 should be used for production.

[0059] According to the present invention, while the mold 40 is fixed to the clamping device 30, it is checked whether or not the mold 40 should be used for production. If, due to human error, the wrong mold 40 is fixed to the clamping device 30, the image analysis unit 52 detects the error. This detection prevents the wrong mold 40 from being attached to the clamping device 30. When the correct mold 40 is fixed to the clamping device 30, it becomes possible to link this mold 40 with the molding conditions corresponding to this mold 40. Therefore, according to the present invention, a technology is provided that prevents the incorrect mold 40 from being attached to the mold clamping device 30 and also links and manages the mold 40 and molding conditions (see Figure 7).

[0060] Furthermore, in this invention, the image included in the description of discriminating visual features by image recognition may be an image obtained by capturing with an imaging mechanism (original image), or an image obtained by, for example, adjusting the magnification of the original image (another image). In other words, the image may be the original image alone, the other image alone, or (original image + other image). Also, the other image may be multiple different images.

[0061] From this perspective, the present invention can be summarized as follows. As shown in Figure 1(a), the process involves preparing an imaging mechanism 50 that photographs the top surface or a specific surface of the mold 40, and an image analysis unit 52 that analyzes the image captured by the imaging mechanism 50. In the image analysis unit 52, which has the molding conditions corresponding to the mold 40 inputted, a registration number is assigned to the mold 40 as shown in Figure 5 (ST13), a first feature is extracted from the image as shown in Figure 6 (ST151), then the magnification of the image is increased (ST153), a second feature is extracted from the obtained image (ST155), then the magnification of the image is further increased (ST156), a third feature is extracted from the obtained image (ST157), and the obtained first to third features are registered in a ledger (Figure 7) in a feature extraction process (Figure 5). When we refer to the mold selected based on the production plan as mold A, the molding conditions corresponding to mold A as molding conditions B, the image of the top surface or a specific surface of mold A as image C, and the first feature obtained from image C as first feature D1, the second feature as second feature D2, and the third feature as third feature D3, As shown in Figure 8, the molding conditions B corresponding to the mold A are input to the image analysis unit (ST21), the mold A is fixed to the mold clamping device (ST22), the top surface or a specific surface of the mold A is photographed with the imaging mechanism (ST23), the first feature D1 is extracted from the obtained image C (ST24), the magnification of image C is increased and the second feature D2 is extracted from the obtained image (ST24), the magnification of image A is further increased and the third feature D3 is extracted from the obtained image (ST24), the registration number matching the first to third features D1 to D3 is identified from the ledger (ST25), and it is confirmed that the molding conditions associated with this identified registration number match the molding conditions B of the mold A (ST26), in a mold matching step, A method for verifying molds consisting of the above is provided.

[0062] According to the present invention, while the mold 40 is fixed to the clamping device 30, it is checked whether or not the mold 40 should be used for production. If, due to human error, the wrong mold 40 is fixed to the clamping device 30, the image analysis unit 52 detects the error. This detection prevents the wrong mold 40 from being attached to the clamping device 30. When the correct mold 40 is fixed to the clamping device 30, it becomes possible to link this mold 40 with the molding conditions corresponding to this mold 40. Therefore, according to the present invention, a technology is provided that prevents the incorrect mold 40 from being attached to the mold clamping device 30 and also links and manages the mold 40 and molding conditions (see Figure 7).

[0063] The matching process is then performed based on the first, second, and third features. This is a form of matching similar to three-factor authentication, significantly increasing the reliability of the authentication.

[0064] Furthermore, if a decrease in reliability in authentication is acceptable, the following invention is effective. That is, as shown in Figure 1(a), the process involves preparing an imaging mechanism 50 that photographs the top surface or a specific surface of the mold 40, and an image analysis unit 52 that analyzes the image captured by the imaging mechanism 50. The image analysis unit 52, which has the molding conditions corresponding to the mold 40 inputted, performs a feature extraction step (Figure 5) in which, as shown in Figure 5, a registration number is assigned to the mold 40 (ST13), as shown in Figure 6, a first feature is extracted from the image (ST151), the magnification of the image is then increased (ST153), a second feature is extracted from the obtained image (ST154), and the obtained first and second features are registered in a ledger (Figure 7). When a mold selected based on a production plan is called mold A, the molding conditions corresponding to mold A are called molding conditions B, an image of the top surface or a specific surface of mold A is called image C, and the first feature obtained from image C is called first feature D1, and the second feature is called second feature D2, As shown in Figure 8, the molding conditions B corresponding to the mold A are input to the image analysis unit (ST21), the mold A is fixed to the mold clamping device (ST22), the top surface or a specific surface of the mold A is photographed with the imaging mechanism (ST23), the first feature D1 is extracted from the obtained image C (ST24), the magnification of the image C is then increased and the second feature D2 is extracted from the obtained image (ST24), the registration number matching the first and second features D1 and D2 is identified from the ledger (ST25), and it is confirmed that the molding conditions associated with this identified registration number match the molding conditions B of the mold A (ST26), in a mold matching step, A method for verifying molds consisting of the above is provided.

[0065] According to the present invention, while the mold 40 is fixed to the clamping device 30, it is checked whether or not the mold 40 should be used for production. If, due to human error, the wrong mold 40 is fixed to the clamping device 30, the image analysis unit 52 detects the error. This detection prevents the wrong mold 40 from being attached to the clamping device 30. When the correct mold 40 is fixed to the clamping device 30, it becomes possible to link this mold 40 with the molding conditions corresponding to this mold 40. Therefore, according to the present invention, a technology is provided that prevents the incorrect mold 40 from being attached to the mold clamping device 30 and also links and manages the mold 40 and molding conditions (see Figure 7).

[0066] The matching process is then performed based on the first and second features. This is a form of matching similar to two-factor authentication, and the reliability of the authentication is significantly higher than that of single-factor authentication.

[0067] Furthermore, the mold 40 used in the method of the present invention is preferably an injection molding mold that is clamped by the clamping device 30 of the injection molding apparatus 10, as shown in Figure 1(a). To implement the method of the present invention, it is necessary to attach an imaging mechanism 50 and an image analysis unit 52 to a general-purpose injection molding apparatus. Since the image analysis unit 52 performs advanced processing, it is inevitably expensive. Therefore, the cost of the injection molding machine 10 also increases.

[0068] Incidentally, according to one survey, when the total production ratio of molds in Japan is set at 100% (on a monetary basis), resin molding molds account for 38%, press molds for 35%, casting molds including die casting for 10%, and other molds for 17%. In the largest category of resin molding dies, injection molding dies account for the largest proportion.

[0069] Inevitably, the number and types of injection molding dies will increase. When the present invention is applied to such an injection molding die, the increased cost due to the addition of the imaging mechanism 50 and the image analysis unit 52 can be recovered in a short period of time. In other words, from the viewpoint of cost-effectiveness, the method of the present invention is more suitable for injection molding dies than other dies. [Industrial applicability]

[0070] The present invention is suitable for a method of verifying molds fixed to a clamping device of an injection molding apparatus. [Explanation of Symbols]

[0071] 10...Injection molding machine, 30...Clamping machine, 40...Mold, 50...Imaging mechanism, 52...Image analysis unit, 53...Image, 65...Edge, 67...Ledger.

Claims

1. We prepare an imaging mechanism that photographs the top surface or a specific surface of the mold, and an image analysis unit that analyzes the image captured by this imaging mechanism. A mold verification method comprising determining the visual characteristics of the mold by image recognition using the image analysis unit, and verifying whether or not the mold fixed to the mold clamping device should be used for production.

2. A step of preparing an imaging mechanism for photographing the top surface or a specific surface of a mold, and an image analysis unit for analyzing the image taken by this imaging mechanism, The image analysis unit, which has the molding conditions corresponding to the mold inputted, assigns a registration number to the mold, extracts a first feature from the image, then increases the magnification of the image and extracts a second feature from the resulting image, then further increases the magnification of the image and extracts a third feature from the resulting image, and registers the obtained first to third features in a ledger in a feature extraction step. When we refer to the mold selected based on the production plan as mold A, the molding conditions corresponding to mold A as molding conditions B, the image of the top surface or a specific surface of mold A as image C, and the first feature obtained from image C as first feature D1, the second feature as second feature D2, and the third feature as third feature D3, A mold matching step includes inputting the molding conditions B corresponding to the mold A into the image analysis unit, fixing the mold A to the mold clamping device, photographing the top surface or a specific surface of the mold A with the imaging mechanism, extracting the first feature D1 from the obtained image C, then increasing the magnification of image C and extracting the second feature D2 from the obtained image, then further increasing the magnification of image A and extracting the third feature D3 from the obtained image, identifying the registration number from the ledger that matches the first to third features D1 to D3, and confirming that the molding conditions associated with this identified registration number match the molding conditions B of the mold A. A method for verifying molds consisting of the following components.

3. A step of preparing an imaging mechanism for photographing the top surface or a specific surface of the mold, and an image analysis unit for analyzing the image taken by the imaging mechanism, The image analysis unit, which has the molding conditions corresponding to the mold inputted, assigns a registration number to the mold, extracts a first feature from the image, then increases the magnification of the image, extracts a second feature from the obtained image, and registers the obtained first and second features in a ledger in a feature extraction step. When we refer to the mold selected based on the production plan as mold A, the molding conditions corresponding to mold A as molding conditions B, the image of the top surface or a specific surface of mold A as image C, the first feature obtained from image C as first feature D1, and the second feature as second feature D2, A mold matching step involves inputting the molding conditions B corresponding to the mold A into the image analysis unit, fixing the mold A to the mold clamping device, photographing the top surface or a specific surface of the mold A with the imaging mechanism, extracting the first feature D1 from the obtained image C, then increasing the magnification of the image C, extracting the second feature D2 from the obtained image, identifying the registration number from the ledger that matches the first to second features D1 to D2, and confirming that the molding conditions associated with this identified registration number match the molding conditions B of the mold A. A method for verifying molds consisting of the following components.

4. A method for verifying a mold according to any one of claims 1 to 3, A method for verifying a mold, characterized in that the mold is a mold clamped by a clamping device of an injection molding apparatus.

Images