Identification device and identification program

Abstract

To provide an identification apparatus and an identification program which can specify a type of a molding and can identify the molding without any additional processing on a surface of the molding for identification at a manufacturing stage.SOLUTION: A collation irregular image as an image of a region with a collation regular image as an image of a region formed with regular design of the collation molded product and an irregular design that is formed when the regular design is formed and formed in a region different from the region formed with the regular design is acquired. With reference to a storage unit for storing, for each mold, a pair of a registered regular image as an image of a region formed with the regular design of each of a plurality of molding products formed with use of a mold and a registered irregular image as an image of a region formed with the irregular design is set as a registered image, a mold for use in molding the collation molded product is specified by using the registered regular image and the collation regular image. It is discriminated whether or not the collation irregular image is matched with the registered irregular image of any of the registered images corresponding to the specified mold.SELECTED DRAWING: Figure 4

Description

Technical Field

[0001] The present invention relates to an identification device and an identification program.

Background Art

[0002] Patent Document 1 discloses a collation device having acquisition means for setting at least a part of a skeleton portion of an identification mark attached to a component or a product as a collation area, and acquiring a first image of a matte pattern formed in the collation area and a second image of a matte pattern formed in a collation area of a skeleton portion of an identification mark attached to a component or a product to be identified, or a skeleton portion of a component that is one of the components of the product to be identified, and having output means for collating the first image and the second image and outputting the collation result. The matte pattern is formed by a mold and then, as a subsequent process, is formed by blast processing, matte processing by powder coating, or plating. The roughness of the pattern formed by the mold is made rougher than the pattern formed by the subsequent process.

[0003] Patent Document 2 discloses a collating device that includes at least an extraction means for extracting an n-dimensional symbol indicating information related to a component, a product, or a product having the component as a constituent element, and an embossed pattern image formed in a collation area that is at least a part of a predetermined area of a component or a product having an embossed pattern applied by a mold, from a captured image; an acquisition means for acquiring information related to the component or the product from the extracted n-dimensional symbol image; and a means for obtaining and outputting a collation result by collating the extracted embossed pattern image with a database in which image features of the embossed pattern formed in the collation area of at least one of a plurality of components or products having an embossed pattern applied by the same mold are stored as representative values, with at least a part of a predetermined area of a component or a product having an embossed pattern applied by the mold as the collation area. The embossed pattern is one obtained by performing, as a subsequent process, blasting, embossing by powder coating, or plating on the embossed pattern formed by the mold, and by making the roughness of the embossed pattern by the mold coarser than the pattern of individual differences caused by the subsequent process, it is easy to independently identify the mold and the individual differences.

Prior Art Documents

Patent Documents

[0004]

Patent Document 1

Patent Document 2

Summary of the Invention

Problems to be Solved by the Invention

[0005] In the case of a molded product, there are cases where it is desired to identify the mold used to mold the molded product after manufacturing, which is called mold identification. Also, there are cases where it is desired to identify which of the data of individual molded products left as a history corresponds, not only to the molded mold, but also to this, which is called individual identification. For example, after a molded product is manufactured and circulated in the market, in order to trace the manufacturing history, a method of performing individual identification after performing mold identification is used.

[0006] As a method of performing individual identification after mold identification, like the identification device according to Patent Document 1 or Patent Document 2, there is also a method of performing mold identification based on the pattern of the molded product and performing individual identification by satin finishing or the like performed on a specific collation area after molding. However, from the viewpoints of avoiding the complexity of the manufacturing process and the homogeneity of the appearance of the molded product, it is necessary to consider a method in which no additional processing is performed on the molded product after molding.

[0007] An object of the present invention is to provide an identification device and an identification program capable of performing mold identification and individual identification of a molded product without performing additional processing on the surface of the molded product for individual identification at the manufacturing stage.

Means for Solving the Problems

[0008] The identification device according to the first aspect includes a processor, and the processor includes a collation fixed image that is at least a partial image of an area where a fixed pattern is formed on a collation molded product that is a collation target, and a pattern that is formed when the fixed pattern is formed and is formed in an area different from the area where the fixed pattern is formed. An irregular pattern, and a collation irregular image that is at least a partial image of an area where the irregular pattern is formed, and a registration fixed image that is at least a partial image of an area where a fixed pattern is formed for each of a plurality of molded products molded using at least one mold, and an irregular pattern A pair with a registration irregular image that is at least a partial image of the area where the pattern is formed is used as a registration image, and the storage unit stored in advance for each type is referred to, and the mold used to mold the collation molded product is specified using the registration fixed image and the collation fixed image, and the collation irregular image is identified as to which registration irregular image of any of the registration images corresponding to the specified mold matches.

[0009] The identification device according to the second aspect is the identification device according to the first aspect, wherein the processor obtains a pair of the collated regular image and the collated irregular image that are each one image, and refers to the storage unit that stores in advance the registered image that is a pair of the registered regular image and the registered irregular image that are each one image.

[0010] The identification device according to the third aspect is the identification device according to the second aspect, wherein the processor obtains a pair of the collated regular image and the collated irregular image that are captured at an imaging resolution corresponding to a region where a finer pattern is formed, between a region where the regular pattern is formed and a region where the irregular pattern is formed, and refers to the storage unit that stores in advance the registered image that is captured at the imaging resolution.

[0011] The identification device according to the fourth aspect is the identification device according to the first aspect, wherein the processor obtains a pair of the collated regular image and the collated irregular image that are each separate images, and refers to the storage unit that stores in advance the registered image that is a pair of the registered regular image and the registered irregular image that are each separate images.

[0012] The identification device according to the fifth aspect is the identification device according to the fourth aspect, wherein the processor obtains a pair of the collated regular image that is captured at a resolution suitable for the region where the regular pattern is formed and the collated irregular image that is captured at a resolution suitable for the region where the irregular pattern is formed, and refers to the storage unit that stores in advance the registered image that is a pair of the registered regular image that is captured at a resolution suitable for the region where the regular pattern is formed and the registered irregular image that is captured at a resolution suitable for the region where the irregular pattern is formed.

[0013] The identification device according to the sixth aspect is the identification device according to the fifth aspect, wherein the processor acquires a pair of the collated shaped image and the collated amorphous image captured by imaging the region where the amorphous pattern is formed with a second resolution higher than the first resolution when imaging the region where the shaped pattern is formed, and refers to the storage unit that stores in advance the registered image that is a pair of the registered shaped image captured at the first resolution and the registered amorphous image captured at the second resolution.

[0014] The identification device according to the seventh aspect is the identification device according to any one of the first to sixth aspects, wherein the amorphous pattern is at least one of a trace of a tool that is used auxiliary in the molding using the mold and a pattern of a region where the shaped pattern is incompletely formed.

[0015] The identification program according to the eighth aspect causes a computer to acquire a collated shaped image that is at least a partial image of a region where a shaped pattern is formed on a collated molded product that is a target of collation, and a collated amorphous image that is at least a partial image of a region where an amorphous pattern is formed, the amorphous pattern being a pattern formed when the shaped pattern is formed and formed in a region different from the region where the shaped pattern is formed, refer to a storage unit that stores in advance, for each mold, a pair of a registered shaped image that is at least a partial image of a region where a shaped pattern is formed on each of a plurality of molded products molded using at least one mold and a registered amorphous image that is at least a partial image of a region where an amorphous pattern is formed as a registered image, specify the mold that molded the collated molded product using the registered shaped image and the collated shaped image, and identify whether the collated amorphous image matches the registered amorphous image of any of the registered images corresponding to the specified mold.

Advantages of the Invention

[0016] According to the first aspect and the eighth aspect, there is an effect that it is possible to provide an identification device and an identification program capable of specifying a mold of a molded product and identifying an individual without performing additional processing on the surface of the molded product for individual identification in the manufacturing stage.

[0017] According to the second aspect, there is an effect that image acquisition is performed more simply as compared with the case of using pairs of a collation template image and a collation non-template image that are separate images, and pairs of a registration template image and a registration non-template image that are separate images.

[0018] According to the third aspect, there is an effect that finer patterns are discriminated as compared with the case of acquiring images without considering the imaging resolution between the collation template image and the collation non-template image, and between the registration template image and the registration non-template image.

[0019] According to the fourth aspect, there is an effect that an imaging region corresponding to the characteristics of the template image and the non-template image is selected as compared with the case of using pairs of a collation template image and a collation non-template image that are one image, and pairs of a registration template image and a registration non-template image that are one image.

[0020] According to the fifth aspect, there is an effect that an imaging resolution corresponding to the characteristics of each of the template image and the non-template image is selected as compared with the case of acquiring pairs of a collation template image and a collation non-template image, and pairs of a registration template image and a registration non-template image at the same imaging resolution.

[0021] According to the sixth aspect, there is an effect that finer non-template patterns are discriminated as compared with the case of using an image pair captured by setting a second resolution for imaging a region where a non-template pattern is formed to be lower than a first resolution for imaging a region where a template pattern is formed.

[0022] According to the seventh aspect, there is an effect that it is easier to specify the region of the non-template pattern as compared with the case of selecting the region of the non-template pattern without considering the formation process.

Brief Description of the Drawings

[0023]

Figure 1

Figure 2

Figure 3

Figure 4

Figure 5

Figure 6

Figure 7

Figure 8

Figure 9

Mode for Carrying Out the Invention

[0024] Hereinafter, with reference to the drawings, embodiments for carrying out the present invention will be described in detail. In the following description, an identification device and an identification program according to the present invention will be exemplified and described in a form applied to an identification device and an identification program for specifying a mold and identifying an individual after manufacturing, for a molded product formed by a mold. Here, the "mold" in the present embodiment means a mold for forming a regular pattern on many parts of the surface of a molded product for molding, and examples of the mold include a die-casting method, an injection molding method, and die forging. A product molded using the mold is called a "molded product". In the following description, the surface opposite to the back surface is referred to as the "front surface", and the appearance surface regardless of the front and back is referred to as the "surface".

[0025] [First Embodiment] With reference to FIGS. 1 to 7, an identification device and an identification program according to the present embodiment will be described. As shown in FIG. 1, the identification device 10 according to the present embodiment includes a CPU (Central Processing Unit) 11, a ROM (Read Only Memory) 12, a RAM (Random Access Memory) 13, an input unit 15, a UI (User Interface) unit 16, and a storage unit 17. Note that the CPU 11 is an example of the "processor" according to the present invention.

[0026] The CPU 11 overall controls the identification device 10 and executes an identification process for specifying a mold and identifying an individual according to the present embodiment. The ROM 12 stores various programs including a control program of the identification device 10, an identification process program executed by the identification device 10, and data. The RAM 13 is a memory used as a work area when various programs are executed. The CPU 11 performs an identification process by expanding and executing the program stored in the ROM 12 in the RAM 13.

[0027] The input unit 15 (reception unit) inputs (receives) an image used when performing identification processing on a molded product to be collated. In the present embodiment, in the manufacturing stage, an image of a specific region on the surface of the molded product is captured for each molded product by a predetermined method and stored in the storage unit 17 as a registered image. Further, for example, in order to perform mold identification and individual identification on a certain molded product returned from the market for tracking the manufacturing history (manufacturing date, manufacturing lot, equipment used, etc.) as a collation molded product, a collation image obtained by imaging the surface of a specific region of the collation molded product by the same predetermined method as at the time of manufacturing is used. The input unit 15 is a part that receives these registered images and collation images. The input unit 15 may be an interface such as a USB (Universal Serial Bus) or a communication interface that inputs the registered image and the collation image as data, or may be an imaging device such as a camera.

[0028] The UI unit 16 is a part where the user gives instructions such as execution of identification processing, input of registered images, and collation images. The UI unit 16 is, for example, a liquid crystal monitor equipped with a touch panel function.

[0029] The storage unit 17 mainly stores the above-mentioned registered images. The storage unit 17 is, for example, an HDD (Hard Disk Drive), an SSD (Solid State Drive), or a flash memory. Note that the storage unit 17 may store an identification processing program or the like.

[0030] As shown in FIG. 1, each of the CPU 11, the ROM 12, the RAM 13, the input unit 15, the UI unit 16, and the storage unit 17 is interconnected by a bus 18.

[0031] Here, in general products, it may be necessary to track the manufacturing history after the products are manufactured and put on the market. Even in molded products manufactured using a mold, there may be a demand to track the manufacturing history of the operating molded products, and in that case, it is first necessary to identify the mold used. In many parts of the surface of the molded product, a mold-specific texture pattern (fixed pattern, for example, a matte finish) may be formed, and by using this fixed pattern, the mold can be identified. However, for tracking the manufacturing history, it is further necessary to identify (individually identify) the molded product as an individual formed by that mold. However, as will be described later, the fixed pattern is not a pattern that can stably perform individual identification.

[0032] On the other hand, in the mold forming process, traces of auxiliary tools such as ejector pins may be formed on the molded product, or incomplete parts of the fixed pattern may be formed even without using auxiliary tools. An ejector pin is a pin used to remove the molded product from the mold. These traces are uncontrollable in the mold forming process (for example, the end face of an ejector pin with a unique pattern may rotate and result in different traces for each molding). That is, when using auxiliary tools, there is a high possibility that an irregular pattern will be formed on the surface of the molded product. This embodiment is based on the finding that by using this irregular pattern, mold identification and individual identification of the molded product can be performed in the manufacturing stage without performing additional processing on the surface of the molded product for individual identification.

[0033] Therefore, in the present invention, in the manufacturing stage, a pair of images (registered images) of the shaped pattern area (shaped image) and the unshaped pattern area (unshaped image) of the molded product are stored in the storage unit. When identifying the mold and individual after manufacturing, the collated shaped image of the collated molded product is compared with the registered shaped image to first identify the mold that molded the collated molded product. Next, the collated unshaped image is compared with the registered unshaped image associated with the identified mold to uniquely identify the registered image corresponding to the collated molded product. By doing so, the mold identification and individual identification of the molded product are performed without performing additional processing on the surface of the molded product for individual identification in the manufacturing stage. Also, after identifying one of the plurality of molds, by comparing the registered image associated with the identified mold with the collated image, it is not necessary to compare all the images related to all the molds with the captured image of the collated molded product. As a result, even when it is necessary to increase the resolution of the unshaped pattern image in order to perform individual identification more accurately, the computational load is reduced (specifically, 1 / number of molds).

[0034] Here, the "shaped pattern" and "unshaped pattern" according to the present embodiment will be described in more detail. The "shaped pattern" according to the present embodiment is a pattern formed on a molded product by a mold in the manufacturing process, and refers to a pattern that cannot be intentionally reproduced by a human. Due to design reasons or the like, it is a ground pattern that occupies most of the surface of the molded product. For example, the pattern on the surface of the mold itself is transferred to the molded product by forming the molded product using the mold, and the pattern on the surface of the mold itself is formed on the molded product. Specifically, among the matte patterns on the surface of the molded product formed by the mold, the portion where the pattern on the surface of the mold itself is transferred corresponds to the shaped pattern. Also, patterns such as QR codes (registered trademarks) that can be intentionally reproduced are not included in the shaped pattern.

[0035] The "irregular pattern" according to one embodiment of the present invention is a pattern formed when a regular pattern is formed, and is a pattern in which the pattern on the surface of the mold itself is transferred to the molded product in an incomplete state for some reason. It is also a pattern formed in an area different from the area where the regular pattern is formed, and refers to a pattern that cannot be reproduced intentionally by humans. For example, in molding using a mold, it is a trace remaining on the surface of the molded product of a tool that is used auxiliaryly for mechanical reasons. Examples of auxiliary tools include ejector pins, gates, runners, and the like. Also, patterns such as QR codes (registered trademarks) that can be intentionally reproduced are not included in the irregular patterns.

[0036] Next, with reference to FIG. 2, the registered image according to the present embodiment and the method of storing the registered image in the storage unit 17 will be described.

[0037] As shown in FIG. 2(b), the registered image includes a pair of a regular image (registered regular image) and an irregular image (registered irregular image). As described above, in the present embodiment, the regular image refers to an image of the area of the regular pattern of the molded product, and the irregular image refers to an image of the area of the irregular pattern of the molded product.

[0038] The storage unit 17 stores registered images of a plurality of molded products formed using each of one or more molds (exemplified as "molds" in FIG. 2(a)) of a certain molded product. In FIG. 2(a), for M molds (mold 1, mold 2, ···, mold M) of a certain molded product, N1 registered images (registered image 1, registered image 2, ···, registered image N1) are stored for mold 1, N2 registered images (registered image 1, registered image 2, ···, registered image N2) are stored for mold 2, and NM registered images (registered image 1, registered image 2, ···, registered image NM) are stored for mold M. Various information (for example, manufacturing history, etc.) about the molded product corresponding to each registered image may be associated.

[0039] Referring to FIG. 3, a specific example of a registered image will be described. FIG. 3 shows an example of a molded product of synthetic resin molded by injection molding using a mold. FIG. 3(a)<1> shows the pattern on the front surface of a sample of the molded product, and <2> shows the pattern on the back surface.

[0040] FIG. 3(a)<1> is an example of an image obtained by imaging (hereinafter sometimes referred to as "shot") the matte pattern (pattern) on the front surface of one molded product twice. As shown in this example, the matte pattern on the front surface is reproduced with good reproducibility to the extent that it can be distinguished by the naked eye even when the shot is changed.

[0041] FIG. 3(a)<2> shows the back surface images of two molded products molded with one mold. As shown in FIG. 3(a)<2>, circular ejector pin traces remain on the back surface. Hereinafter, such a trace area is referred to as "amorphous area 31". On the other hand, the area around the amorphous area 31 is referred to as "shaped area 30". The shaped area 30 is a transfer area of the surface of the mold. Also, the pattern formed in the amorphous area 31 is referred to as "amorphous pattern AP", and the pattern formed in the shaped area 30 is referred to as "shaped pattern FP". The amorphous patterns AP of the ejector pins formed on the back surfaces of the two molded products shown in FIG. 3(a)<2> are clearly different. On the other hand, it can be seen that the two shaped patterns FP are transferred even to fine parts and are similar to each other.

[0042] FIG. 3(b) shows two shots of the back surface of a molded product by a mold different from that of FIG. 3(a). Also in this example, it can be seen that the amorphous area 31, which is a circular ejector pin trace, has a different pattern (amorphous pattern AP) from the shaped area 30. It can also be seen that the shaped pattern FP is transferred even to fine parts.

[0043] Here, the imaging region 32 according to the present embodiment will be described. In the present embodiment, as shown in FIG. 3(b), at least a part of the regular region 30 (indicated by a dotted line in FIG. 3(b)) and at least a part of the irregular region 31 (indicated by a solid line in FIG. 3(b)) are included in one shot to set the imaging region 32. By doing this, the regular pattern FP and the irregular pattern AP can be included in one image. The image obtained by imaging the imaging region 32 corresponds to the registration image shown in FIG. 2(b), the regular pattern FP in the imaging region 32 corresponds to the regular image, and the irregular pattern AP in the imaging region 32 corresponds to the irregular image.

[0044] The resolution when imaging the imaging region 32 is set to, for example, the resolution of the image that requires more detailed information among the regular pattern FP and the irregular pattern AP (for example, about 3000 dpi). By doing this, type identification and individual identification can be performed more accurately. In the present embodiment, a form in which the regular pattern FP and the irregular pattern AP are included in one image is exemplified and described, but it is not limited to this, and a form in which the regular pattern FP and the irregular pattern AP are separate images may also be used. By doing this, the capacity of the image data can be made smaller, and the selection of the regions for imaging the regular pattern FP and the irregular pattern AP respectively becomes more flexible.

[0045] Next, with reference to FIG. 4, the identification process executed by the identification device 10 according to the present embodiment will be described. FIG. 4 is a flowchart showing the flow of the process of the identification process program executed by the identification device 10. This identification process program is stored, for example, in the ROM 12 shown in FIG. 1, and the CPU 11 reads it from the ROM 12 and expands it in the RAM 13 or the like for execution. In the present embodiment, a form in which the CPU 11 executes the identification process is exemplified and described, but it is not limited to this, and an ASIC (Application Specific Integrated Circuit) or dedicated software or the like may be used to configure a device dedicated to the identification process and execute it under the management of the CPU 11. In the following description, it is assumed that a plurality of registration images related to a plurality of types are already stored in the storage unit 17, and a composite molded product to be the target of the identification process is prepared.

[0046] As shown in FIG. 4, a collation image is acquired in step S100. The "collation image" is an image of a collated molded product in the same imaging area 32 as the registered image, and includes a shaped image (collation shaped image) that captures the shaped pattern FP and an unshaped image (collation unshaped image) that captures the unshaped pattern AP, similar to the registered image.

[0047] In step S102, a mold identification process is executed. The mold identification process is performed by comparing and determining the shaped image (registered shaped image) of the registered image with the shaped image (collation shaped image) of the collation image. For example, Yoctrace or the like is used for the algorithm of this determination. This comparison is performed for all types. In the mold identification process, the registered shaped image to be compared with the collation shaped image is selected from any one of the registered shaped images belonging to each type. That is, for example, in the case of the example shown in FIG. 2(a), for mold 1, any registered shaped image from registered image 1 to registered image N1 may be compared with the collation shaped image. The same applies to molds 2 to M. Also, separately from the registered images of each mold, a registered image dedicated to mold identification may be stored.

[0048] In step S104, it is determined whether the mold identification process has ended for all types. If the determination is negative, the process returns to step S102 to continue the mold identification process. If the determination is positive, the process proceeds to step S106.

[0049] In step S106, it is determined whether there is a type that includes a registered shaped image that matches the collation shaped image of the collated molded product. If the determination is positive, the process proceeds to step S108. If the determination is negative, this identification processing program ends. When it is determined that there is a matching type, the mold that molded the collated molded product is identified.

[0050] In step S108, individual identification processing is executed. The individual identification processing is performed by comparing and determining the amorphous image of the registered image (registered amorphous image) and the amorphous image of the comparison image (comparison amorphous image). For example, Yoctrace or the like is used for this determination algorithm. This comparison is performed for all registered images registered in the specified type. For example, when the mold 1 is specified in the example shown in FIG. 2(a), the comparison amorphous image is compared with all the registered amorphous images from the registered image 1 to the registered image N1. Here, when there are a plurality of registered images equal to or higher than the determination threshold as a result of the determination, for example, the registered amorphous image with the highest similarity is adopted. The details of the "determination threshold" and "similarity" will be described later.

[0051] In step S110, it is determined whether the comparison with the comparison amorphous image has been completed for all the registered amorphous images. If the determination is negative, the process returns to step S108 to continue the individual identification processing. If the determination is positive, the process proceeds to step S112.

[0052] In step S112, it is determined whether there is a registered image including the registered amorphous image that matches the comparison amorphous image. If the determination is positive, the process proceeds to step S114. If the determination is negative, the present identification processing program is terminated. When there is a registered amorphous image that matches in step S112, the registered image of the comparison molded product captured is specified, that is, individual identification is performed. In the present embodiment, the case where the determination is negative in step S112 and there is no matching registered amorphous image and the present identification processing is terminated is illustrated and described, but the present invention is not limited thereto, and the registered amorphous image with the highest similarity may be presented as a candidate.

[0053] In step S114, information related to the comparison molded product that has been individually identified (for example, manufacturing history, etc.) is output, and then the present identification processing program is terminated.

[0054] Next, with reference to FIG. 5, the principle of the mold identification process according to the present embodiment will be described. FIG. 5 is a diagram for explaining that mold identification is possible, taking the shaped pattern FP (mat surface) on the front surface of a molded product by aluminum die casting as an example. The horizontal axis in FIG. 5 indicates the similarity, and the vertical axis indicates the frequency (number of shots). The similarity indicates the degree to which images resemble each other and is measured by pattern matching or the like. In FIG. 5, the set indicated by <1> shows the comparison result between others, and the set indicated by <3> shows the comparison result between the same person. "Comparison between others" is a comparison between images of different objects (images of different regions or images of molded products by different molds), and "comparison between the same person" is a comparison between the same images (images by the same shot in the same region). <2> is a comparison between images of different shots in the same region (different molded products by the same mold). The resolution of each image is, as an example, about 600 dpi and 48 - 96 px.

[0055] As shown in FIG. 5, the images included in <1> are clearly different to the extent that they can be judged by visual inspection and are distributed near the minimum value of the similarity. The images included in <3> are very similar and are distributed near the maximum value of the similarity. As is clear from FIG. 5, the set of <1> and the set of <2> are clearly separated, and a determination range th1 with a width that causes no practical problems is set. That is, it can be seen that mold identification is possible from the comparison between <1> and <3>. In the present embodiment, the boundary for distinguishing between two distributions is defined as the "determination threshold value", the range where the frequency between the two distributions becomes 0 is defined as the range where the determination threshold value is set, and this range is defined as the "determination range th".

[0056] On the other hand, since the images included in <2> are very similar and difficult to distinguish, they are distributed in a range where the similarity is relatively high although lower than the set shown by <3>. Therefore, the set shown by <2> and the set shown by <3> are very close. For this reason, the determination range th2 for distinguishing between <2> and <3> has a narrow width, and it is impossible to set the determination range at a level that causes no practical problems, or the tails overlap and thus it is impossible to set it in the first place. That is, when using the mat pattern on the front surface, it can be seen that mold identification is possible, but individual identification is difficult.

[0057] Next, referring to FIG. 6, the principle of individual identification will be described. FIG. 6(a) shows a collation result using an image including a pair of a fixed pattern FP and an amorphous pattern AP within a specific imaging region 32 of a plurality of aluminum die-cast molded products by a certain mold as shown in FIG. 6(b). In the example shown in FIG. 6, since the fixed pattern FP adjacent to the amorphous pattern AP is a relatively distinct pattern, it is treated as a fixed pattern in the same way as the surface texture pattern of the front side shown in FIG. 5. In FIG. 6(a), the set indicated by <1> shows the result of comparing the amorphous patterns AP of different persons, and the set indicated by <2> shows the result of comparing the amorphous patterns AP of the same person. In the example shown in FIG. 6, since the fixed pattern FP and the amorphous pattern AP are adjacent, the image to be collated is regarded as one image. However, when the fixed pattern FP and the amorphous pattern AP are separated, they should be regarded as separate images and linked to each other.

[0058] As shown in <1> of FIG. 6, the amorphous patterns AP of different persons are clearly different from each other, and the similarity degrees are distributed near the minimum value. On the contrary, as shown in <2>, the amorphous patterns AP of the same person are similar to each other, and the similarity degrees are distributed near the maximum value. Therefore, a determination range th3 without practical problems is set between the group of comparing different persons shown in <1> and the group of comparing the same person shown in <2>. From this, it can be seen that individual identification can be accurately performed by using the amorphous pattern AP. Here, as the pattern on the surface of the molded product used as the amorphous pattern AP, in addition to the ejector pin described above, for example, a defective release agent application part (a part where the application of the release agent has shading), a gate part, etc. can be mentioned. The amorphous pattern AP shown in FIG. 6 is an example using a gate part.

[0059] Note that the image shown in FIG. 6(b) is captured at a resolution of 1000 dpi as an example. By setting the resolution to this level, the uneven portions of the amorphous pattern AP can be captured in detail. On the other hand, for the fixed pattern FP, details can be captured even at a resolution of about 300 dpi. Therefore, when the fixed pattern FP and the amorphous pattern AP are separate images, different resolutions such as 300 dpi for the former and 1000 dpi for the latter may be used. This saves the capacity of the storage unit 17 when storing the images.

[0060] Next, with reference to FIG. 7, the imaging region 32 according to the present embodiment will be described in more detail.

[0061] The examples of <1>, <2>, and <3> shown in FIG. 7(a) all show patterns formed by three different molds using substantially circular ejector pins. In each pattern, the ejector pin forms a unique trace, and in the region adjacent to the ejector pin, a fine pattern comparable to the front surface pattern is formed. Therefore, as shown in FIG. 7(a), an imaging region 32 including a fixed pattern FP and an amorphous pattern AP is set.

[0062] FIG. 7(b) shows an example of the amorphous pattern AP when a gate is used as the amorphous region 31 (denoted as "amorphous surface" in FIG. 7(b)). The "gate" is the inlet of the resin, which is the material of the molded product. <1> and <2> shown in FIG. 7(b) show images of the gate portions of different molded products of the same mold. As shown in FIG. 7(b), even when the same mold is used, the amorphous patterns AP of the gate portions are different enough to be sufficiently distinguished.

[0063] FIG. 7(c) shows an example of the amorphous pattern AP when the vicinity of the runner is used as the amorphous region 31 (denoted as "amorphous surface" in FIG. 7(b)). The "runner" is the passage through which the resin, which is the material of the molded product, flows. <1> and <2> shown in FIG. 7(c) show images of the runner portions of different molded products of the same mold. As shown in FIG. 7(c), even when the same mold is used, the amorphous patterns AP of the runner portions are different enough to be sufficiently distinguished.

[0064] [Second Embodiment] With reference to FIG. 8, the identification device and identification program according to the present embodiment will be described. In the above embodiment, as an amorphous pattern, a form using the traces of auxiliary tools used for molding was exemplified and described. However, in the present embodiment, as the amorphous pattern, a form using a portion where the surface of the mold (either the front surface or the back surface) is incompletely transferred is used. Therefore, since the identification device itself is the same as in the above embodiment, if necessary, FIGS. 1, 2, and 4 may be referred to, and detailed description will be omitted.

[0065] The present embodiment is a form applied to a cap molded by injection molding using synthetic resin as shown in FIG. 8(b) as a molded product. In this example of the cap, there is a region on the back side of the cap where the surface of the mold is not completely transferred (hereinafter sometimes referred to as an "incomplete transfer region"). The present embodiment uses the pattern of this incomplete transfer region as the amorphous pattern AP.

[0066] The distribution shown by <1> in FIG. 8(a) shows the result of comparison between different people for a plurality of caps shown in FIG. 8(b), and the distribution shown by <2> shows the result of comparison between images of different shots (different molded products by the same mold) in the same region. As shown by <1> in FIG. 8(a), the result of comparison between different people in this example is distributed in the vicinity of a relatively low similarity. On the other hand, the distribution shown by <2> has a wider distribution and a longer tail than the distribution shown by <1>. However, in this example, a determination range th4 for distinguishing between the distribution shown by <1> and the distribution shown by <2> is set. That is, individual identification is performed using the determination range th4.

[0067] As described above, in this example, a determination range th4 is set, but the width of the determination range th4 is narrow, and the possibility of causing an incorrect determination in individual identification cannot be eliminated. That is, the similarity may decrease due to slight scratches or dirt, and in that case, the distributions of <1> and <2> may overlap. Even in such a case, by performing the identification process shown in FIG. 4, incorrect determination is suppressed. That is, when there are a plurality of molds for molding the cap of this example, first, the collated shaped image and the registered shaped image (FIG. 2) are compared to perform mold identification processing to identify the molded mold. Next, the collated indeterminate image is compared with the registered indeterminate image (FIG. 2) associated with the identified mold to perform individual identification processing. As a result, since the registered indeterminate image to be compared with the collated indeterminate image is limited, incorrect determination in individual identification is suppressed.

[0068] [Third Embodiment] Referring to FIG. 9, the identification device and the identification program according to this embodiment will be described. This embodiment is a form using a collective molding die that collectively molds a plurality of molded products with one die. Therefore, since the identification device itself is the same as that of the above embodiment, FIGS. 1, 2, and 4 will be referred to if necessary, and detailed description will be omitted.

[0069] FIG. 9(a) shows the appearance of a collective molding die 20 that collectively molds the lid 22 shown in FIG. 9(b). The collective molding die 20 is a die that molds a plurality of lids 22 at once by injection molding using synthetic resin as an example. As shown in FIG. 9(a), the collective molding die 20 includes a plurality (32 in the example of FIG. 9(a)) of die parts 21, and each of the die parts 21 molds the lid 22 shown in FIG. 9(b). In this embodiment, as an example, similar to the second embodiment above, the pattern of the incomplete transfer region is used as an indeterminate pattern. Of course, like the first embodiment, the traces of some auxiliary tools used in molding may be used as the indeterminate pattern.

[0070] In the identification device and the identification program according to the present embodiment, the identification process shown in FIG. 4 is also executed. That is, for example, when there is one integrated molding die 20, the type identification process is performed by comparing the collated shaped image and the registered shaped image. The type identification process in this case is a process of identifying any one of the 32 type portions 21. Next, the individual identification process is performed by comparing the registered indefinite image and the collated indefinite image associated with the identified type portion 21. As a result, the computational load is reduced to 1 / 32 compared to the case of comparing with the registered images associated with all the type portions 21. In addition, since the registered indefinite images to be collated are narrowed down, misjudgment is suppressed for the same reason as in the second embodiment.

[0071] In the above embodiment, the processor refers to a processor in a broad sense, and includes a general-purpose processor (for example, CPU: Central Processing Unit, etc.) and a dedicated processor (for example, GPU: Graphics Processing Unit, ASIC: Application Specific Integrated Circuit, FPGA: Field Programmable Gate Array, programmable logic device, etc.). Further, the operation of the processor in the above embodiment may be achieved not only by one processor but also by a plurality of physically separated processors cooperating with each other. Also, the order of each operation of the processor is not limited to the order described in the above embodiment and may be changed as appropriate.

Explanation of Signs

[0072] 10 Identification device 11 CPU 12 ROM 13 RAM 15 Input unit 16 UI unit 17 Storage unit 18 Bus 20 Integrated molding die 21 Type portion 22 Lid 30 Shaped area 31 Indefinite area 32 imaging areas FP fixed pattern AP amorphous pattern th1, th2, th3, th4 determination ranges

Claims

1. comprising a processor, wherein the processor obtains a collated shaped image which is an image of at least a part of a region where a shaped pattern is formed on a shaped article to be collated, and a collated unshaped image which is an image of at least a part of a region where an unshaped pattern is formed, the unshaped pattern being formed in a region different from the region where the shaped pattern is formed, when the shaped pattern is formed; refers to a storage unit that stores, in advance for each mold, a pair of a registered shaped image which is an image of at least a part of a region where a shaped pattern is formed on each of a plurality of shaped articles molded using at least one mold, and a registered unshaped image which is an image of at least a part of a region where an unshaped pattern is formed, as a registered image; identifies the mold that molded the collated shaped article using the registered shaped image and the collated shaped image, and determines whether the collated unshaped image matches the registered unshaped image of any of the registered images corresponding to the identified mold; the shaped pattern is a part of a surface texture pattern of a shaped article formed by a mold, where the pattern on the surface of the mold itself is transferred, and the unshaped pattern is a pattern transferred to the shaped article in a state where the pattern on the surface of the mold itself is incomplete; an identification device.

2. wherein the processor obtains a pair of the collated shaped image and the collated unshaped image, which are made into one image by imaging such that at least a part of the region where the shaped pattern is formed and at least a part of the region where the unshaped pattern is formed are included in one shot; refers to the storage unit that stores in advance the registered image, which is a pair of the registered shaped image and the registered unshaped image, which are made into one image by imaging such that at least a part of the region where the shaped pattern is formed and at least a part of the region where the unshaped pattern is formed are included in one shot; The identification device according to claim 1.

3. wherein the processor Obtain a pair of the collated shaped image and the collated unshaped image captured at an imaging resolution corresponding to a region where a finer pattern is formed, between the region where the shaped pattern is formed and the region where the unshaped pattern is formed. Refer to a storage unit that has previously stored the registered image captured at the imaging resolution. The identification device according to claim 2.

4. The processor Obtain a pair of the collated shaped image and the collated unshaped image that are made into separate images by separately imaging. Refer to the storage unit that has previously stored the registered image, which is a pair of the registered shaped image and the registered unshaped image that are made into separate images by separately imaging. The identification device according to claim 1.

5. The processor Obtain a pair of the collated shaped image captured at a resolution suitable for the region where the shaped pattern is formed and the collated unshaped image captured at a resolution suitable for the region where the unshaped pattern is formed, and refer to the storage unit that has previously stored the registered image, which is a pair of the registered shaped image captured at a resolution suitable for the region where the shaped pattern is formed and the registered unshaped image captured at a resolution suitable for the region where the unshaped pattern is formed. The identification device according to claim 4.

6. The processor Obtain a pair of the collated shaped image and the collated unshaped image captured by setting a second resolution for imaging the region where the unshaped pattern is formed to be higher than a first resolution for imaging the region where the shaped pattern is formed, and refer to the storage unit that has previously stored the registered image, which is a pair of the registered shaped image captured at the first resolution and the registered unshaped image captured at the second resolution. Refer to the storage unit that has previously stored the registered image, which is a pair of the registered shaped image captured at the first resolution and the registered unshaped image captured at the second resolution. The identification device according to claim 5.

7. The amorphous pattern is at least one of the traces of the tool that is used as an auxiliary in the molding using the mold and the pattern of the region where the shaped pattern is incompletely formed. The identification device according to any one of claims 1 to 6.

8. To a computer, An alignment shaped image which is an image of at least a part of the region where the shaped pattern is formed of the alignment molded product which is the molded product to be aligned, and a pattern formed when the shaped pattern is formed, and is different from the region where the shaped pattern is formed An alignment amorphous image which is an image of at least a part of the region where the amorphous pattern formed in the region is formed, Referring to a storage unit stored in advance for each mold, as a registration image, a pair of a registration shaped image which is an image of at least a part of the region where the shaped pattern is formed of each of a plurality of molded products molded using at least one mold, and a registration amorphous image which is an image of at least a part of the region where the amorphous pattern is formed, Identifying the mold that molded the alignment molded product using the registration shaped image and the alignment shaped image, and identifying whether the alignment amorphous image matches the registration amorphous image of any of the registration images corresponding to the identified mold, The shaped pattern is a part of the surface texture pattern of the molded product formed by the mold, where the pattern on the surface of the mold itself is transferred, and the amorphous pattern is a pattern transferred to the molded product in an incomplete state of the pattern on the surface of the mold itself. An identification program for causing the above to be executed.

Images