JP6777890B2 - Object identifier extractor - Google Patents

Description

The present invention relates to an object identifier extraction device, an object identifier extraction method, an identification device, an identification method, an object identifier registration device, an object identifier registration method, an identification collation device, an identification collation method, an object identifier management system, an object identifier management method, and a program. Regarding.

A wide variety of object identification techniques such as barcodes, serial numbers, and IC tags have been proposed or put into practical use as object identification techniques for identifying the identity of objects such as industrial products and commodities. As a kind of such object identification technology, a pattern containing minute particles is generated on an object, a feature amount is extracted from an image obtained by imaging the pattern, and the extracted feature amount is used as an object identifier. There is a technique to do.

For example, a taggant (tracking additive) is used as a minute grain, a pattern containing the taggant is generated on an object, and a predetermined region of the pattern is imaged and extracted from an image obtained. It is described in Patent Document 1, Patent Document 2 and Patent Document 3 that the distribution of grains in a region is used as a feature amount (object identifier of an object).

Specifically, in Patent Document 1, a printing ink or the like mixed with taggant is applied to all or a part of the surface of the reference object to form a taggant distribution pattern, and the taggant distribution pattern is extracted from the read image. The feature quantity is used as the object identifier of the reference object. When collating with the target object, the reading direction, position, and range are set to the same conditions as the reading of the reference object, the image of the tagant distribution pattern on the target object is read, and the feature amount extracted from the image is used as the reference object. Compare with the object identifier of.

Further, in Patent Document 2, an origin as a reference for alignment is set on each object, and the right direction from the origin is defined as the positive direction of the X axis and the downward direction is defined as the positive direction of the Y axis. A feature amount is extracted from a predetermined area, and the extracted feature amount is used as an object identifier of the object. Here, as the shape of the predetermined region, a logo having a rectangular shape, a circular shape, an elliptical shape, a polygonal shape, or another special shape is exemplified.

Further, in Patent Document 3, a reference portion (alignment mark, line, company logo, frame, edge of the object itself or a combination of edges, etc.) is provided on the object, and at least the reference portion is completely covered with a taggant. Spray paint etc. to attach a random pattern. Then, a feature amount depending on the distribution of the taggant is extracted from the image of the reference portion, and it is used as the object identifier of the object.

On the other hand, one of the biometric authentication technologies is iris recognition. In iris recognition, for example, as described in Patent Documents 4 and 5, features are extracted from the iris image of an individual's eye obtained by imaging with a camera, coded, and collated with a pre-registered iris code. And perform personal authentication.

Japanese Unexamined Patent Publication No. 2013-69188 Japanese Unexamined Patent Publication No. 2014-6840 Special Table 2007-534067 Special Table No. 8-504979 U.S. Pat. No. 5,291,560

As described above, there are object authentication and biometric authentication as authentication technologies, but both have formed and developed different technical fields. Therefore, there has been no idea of using the technology cultivated in biometric authentication, especially iris recognition, for authentication of objects such as industrial products and commodities. Moreover, it was difficult to apply the technology cultivated in iris recognition even if an attempt was made to use it for authentication of an object other than the iris. The reason is that the pattern for object identification has a shape that does not resemble an iris, such as a rectangle, a circle, an ellipse, or a polygon.

An object of the present invention is to provide an object identifier extraction device that solves the above-mentioned problem, that is, it is difficult to apply the technology cultivated in iris recognition to the authentication of objects such as industrial products and commodities. It is in.

The object identifier extraction device according to the embodiment of the present invention is
An imaging unit that captures an object with a random pattern on the surface,
An extraction unit that extracts features of an annular image region including the random pattern from the image of the captured object as identification information of the object.
including.

The object identifier extraction method according to another embodiment of the present invention
An image of an object with a random pattern processed on the surface is imaged.
From the image of the captured object, the feature of the annular image region including the random pattern is extracted as the identification information of the object.

The identification device according to another embodiment of the present invention is
Identification is performed using the object identifier extracted by the object identifier extraction device of the present invention.

The collation device according to another embodiment of the present invention is
Collation is performed using the object identifier extracted by the object identifier extraction device of the present invention.

The identification method according to another embodiment of the present invention
Identification is performed using the object identifier extracted by the object identifier extraction method of the present invention.

The collation method according to another embodiment of the present invention
Collation is performed using the object identifier extracted by the object identifier extraction method of the present invention.

The identification collation device according to another embodiment of the present invention is
An imaging unit that captures an object with a random pattern on the surface,
An extraction unit that extracts features of an annular image region including the random pattern from the image of the captured object as identification information of the object.
A determination unit that compares the object identification information with the registered object identification information stored in the storage unit, and determines the object identification / collation based on the comparison result.
including.

The identification collation method according to another embodiment of the present invention
An image of an object with a random pattern processed on the surface is imaged.
From the image of the captured object, the feature of the annular image region including the random pattern is extracted as the identification information of the object.
The identification information of the object is compared with the registered object identification information stored in the storage unit, and the identification / collation of the object is determined based on the comparison result.

Programs according to other embodiments of the present invention
Computer,
An imaging unit that captures an object with a random pattern on the surface,
An extraction unit that extracts features of an annular image region including the random pattern from the image of the captured object as identification information of the object.
To make it work.

The object identification / registration device according to another embodiment of the present invention is
A processing part that processes a random pattern on the surface of an object,
An imaging unit that captures an object with the random pattern processed,
An extraction unit that extracts features of an annular image region including the random pattern from the image of the captured object as identification information of the object.
including.

The object identification registration method according to another embodiment of the present invention
Random pattern is processed on the surface of the object
An image of an object processed with the random pattern is imaged.
From the image of the captured object, the feature of the annular image region including the random pattern is extracted as the identification information of the object.

The object identifier management system according to another embodiment of the present invention is
It has an object identifier registration device and an identification collation device.
The object identifier registration device is
A processing part that processes a random pattern on the surface of an object,
An imaging unit that captures an object with the random pattern processed,
An extraction unit that extracts features of an annular image region including the random pattern from the image of the captured object as identification information of the object.
A registration unit that registers the object identification information in the storage unit as registered object identification information,
Including
The identification collation device is
An imaging unit that captures an object with a random pattern on the surface,
An extraction unit that extracts features of an annular image region including the random pattern from the image of the captured object as identification information of the object.
A determination unit that compares the object identification information with the registered object identification information stored in the storage unit, and determines the object identification / collation based on the comparison result.
including.

The object identifier management method according to another embodiment of the present invention
It has an object identifier registration process and an identification collation process.
In the object identifier registration step,
Random pattern is processed on the surface of the object
An image of an object processed with the random pattern is imaged.
From the image of the captured object, the feature of the annular image region including the random pattern is extracted as the identification information of the object.
The identification information of the object is registered in the storage unit as the identification information of the registered object, and is registered.
In the identification collation step,
An image of an object with a random pattern processed on the surface is imaged.
From the image of the captured object, the feature of the annular image region including the random pattern is extracted as the identification information of the object.
The identification information of the object is compared with the registered object identification information stored in the storage unit, and the identification / collation of the object is determined based on the comparison result.

Since the present invention has the above-mentioned configuration, the technique cultivated in iris recognition can be applied to the certification of objects such as industrial products and commodities.

It is a block diagram of the object identifier extraction apparatus which concerns on 1st Embodiment of this invention. It is a flowchart which shows the procedure of the object identifier extraction method executed by the object identifier extraction apparatus which concerns on 1st Embodiment of this invention. It is a block diagram of the object identifier extraction apparatus which concerns on 2nd Embodiment of this invention. It is a flowchart which shows the procedure of the object identifier extraction method executed by the object identifier extraction apparatus which concerns on 2nd Embodiment of this invention. It is a conceptual diagram of the preprocessing carried out on an image by the object identifier extraction apparatus which concerns on 2nd Embodiment of this invention. It is a block diagram of the object identifier extraction apparatus which concerns on 3rd Embodiment of this invention. It is a flowchart which shows the procedure of the object identifier extraction method executed by the object identifier extraction apparatus which concerns on 3rd Embodiment of this invention. It is a conceptual diagram of the preprocessing carried out on an image by the object identifier extraction apparatus which concerns on 3rd Embodiment of this invention. It is a block diagram of the object identifier extraction apparatus which concerns on 4th Embodiment of this invention. It is a flowchart which shows the procedure of the object identifier extraction method executed by the object identifier extraction apparatus which concerns on 4th Embodiment of this invention. It is a conceptual diagram of the preprocessing carried out on an image by the object identifier extraction apparatus which concerns on 4th Embodiment of this invention. It is a block diagram of the object identifier registration apparatus which concerns on 5th Embodiment of this invention. It is a flowchart which shows the procedure of the object identifier registration method executed by the object identifier registration apparatus which concerns on 5th Embodiment of this invention. It is a block diagram of the object identifier registration apparatus which concerns on 6th Embodiment of this invention. It is a flowchart which shows the procedure of the object identifier registration method executed by the object identifier registration apparatus which concerns on 6th Embodiment of this invention. It is a figure which shows the structural example of the data which the object identifier registration device which concerns on 6th Embodiment of this invention registers in a storage device. It is a block diagram of the identification collation apparatus which concerns on 7th Embodiment of this invention. It is a block diagram of the information processing apparatus which realizes the identification collation apparatus which concerns on 7th Embodiment of this invention. It is a flowchart which shows the procedure of the identification collation method executed by the identification collation apparatus which concerns on 7th Embodiment of this invention. It is a block diagram of the object identifier management system which concerns on 8th Embodiment of this invention. It is a flowchart which shows the procedure of the object identifier management method executed by the object identifier management system which concerns on 8th Embodiment of this invention. It is a block diagram of the object identifier registration apparatus which concerns on 9th Embodiment of this invention. It is a figure which shows the state which the lid is pressed in the object identifier registration device which concerns on 9th Embodiment of this invention. It is a figure which shows the pattern processed on the object by the object identifier registration apparatus which concerns on 9th Embodiment of this invention. It is a block diagram of the pattern image pickup unit in the object identifier registration apparatus which concerns on 9th Embodiment of this invention. It is a flowchart which shows the processing procedure of MPU of the pattern imaging unit in the object identifier registration apparatus which concerns on 9th Embodiment of this invention. It is a block diagram which shows the modification of the object identifier registration apparatus which concerns on 9th Embodiment of this invention. It is a block diagram which shows the modification of the object identifier registration apparatus which concerns on 9th Embodiment of this invention. It is a block diagram of the object identifier registration apparatus which concerns on tenth embodiment of this invention. It is a figure which shows the state which the lid is pressed in the object identifier registration device which concerns on tenth embodiment of this invention. It is a figure which shows the pattern processed on the object by the object identifier registration apparatus which concerns on tenth embodiment of this invention. It is a block diagram of the pattern image pickup unit in the object identifier registration apparatus which concerns on tenth embodiment of this invention. It is a flowchart which shows the processing procedure of MPU of the pattern imaging unit in the object identifier registration apparatus which concerns on tenth embodiment of this invention. It is a block diagram of the object identifier registration apparatus which concerns on 11th Embodiment of this invention. It is a figure which shows the state which the lid is pressed in the object identifier registration device which concerns on 11th Embodiment of this invention. It is a figure which shows the pattern generated on the object by the object identifier registration apparatus which concerns on 11th Embodiment of this invention. It is a block diagram of the pattern image pickup unit in the object identifier registration apparatus which concerns on 11th Embodiment of this invention. It is a flowchart which shows the processing procedure of MPU of the pattern imaging unit in the object identifier registration apparatus which concerns on 11th Embodiment of this invention. It is a block diagram of the object identifier extraction apparatus which concerns on the twelfth embodiment of this invention. It is a flowchart which shows the procedure of the object identifier extraction method executed by the object identifier extraction apparatus which concerns on 12th Embodiment of this invention.

Next, an embodiment of the present invention will be described in detail with reference to the drawings.
[First Embodiment]
Referring to FIG. 1, the object identifier extraction device 100 according to the first embodiment of the present invention has a function of extracting the object identifier 120 of the object 110.

The object 110 is a target for extracting an object identifier. The object 110 is, for example, an industrial product or a product package. Further, the object 110 may be an individual's belongings (business card, notebook, etc.) or a part of the body (for example, a finger) excluding the individual's iris. A random pattern 111 having an annular planar shape is processed on any surface of the object 110. The planar shape is a shape seen from directly above the random pattern 111. Here, it is desirable that the surface of the object 110 on which the random pattern 111 is processed has a color, material, pattern, or the like that is visually distinguishable from the random pattern 111. Further, the size of the random pattern 111 (the size of the circle formed by the outer boundary of the ring) may be about the same as the size of the human iris (the size of the circle formed by the outer boundary of the iris). However, the size is not limited to such a size, and the size of the random pattern 111 may be smaller or larger than the size of the human iris. The same applies to the embodiment described later in which the circular random pattern is used.

The method of processing the random pattern 111 on the surface of the object 110 is arbitrary. For example, the random pattern 111 may be generated by applying a printing ink, a paint, or the like mixed with fine particles to the surface of the object 110 using a pen, a stamp, or the like and solidifying the pattern 110. As the fine particles, fine particles such as metal powder and glass powder, taggants described in Patent Documents 1 to 3 and the like can be used. It is desirable that the fine grains are grains having different reflection characteristics from the material (excluding the above-mentioned fine grains) constituting the random pattern 111. Further, it is desirable that the fine grains are non-uniformly contained in the random pattern 111. Alternatively, the random pattern 111 may be generated by processing a random pattern on the surface of the object 110 with a laser processing machine. Alternatively, when the surface of the object 110 is satin finished, the random pattern 111 may be generated by utilizing the satin finish.

The object identifier extraction device 100 has an imaging unit 101 and an extraction unit 102 as main functional units.

The imaging unit 101 has a function of acquiring an image of the random pattern 111 formed on the object 110. Here, the imaging unit 101 acquires an image in which the planar shape of the random pattern 111 is captured.

The extraction unit 102 has a function of extracting the feature amount of the random pattern 111 from the image acquired by the image pickup unit 101 and outputting the extracted feature amount as the object identifier 120 of the object 110. The extraction unit 102 uses a well-known algorithm (hereinafter, referred to as an iris recognition algorithm) that extracts and encodes features from the iris image in extracting the feature amount.

As shown in FIG. 18, for example, the object identifier extraction device 100 includes an information processing device 740 having an arithmetic processing unit 741 such as one or more microprocessors, a storage unit 742 such as a memory or a hard disk, and a camera 743, and a program. It can be realized with 744. The information processing device 740 may be, for example, a smartphone. The program 744 is read into the memory from an external computer-readable recording medium at the time of starting up the information processing device 740, and controls the operation of the arithmetic processing unit 741 so that the imaging unit 101 is placed on the arithmetic processing unit 741. , The extraction unit 102, and other functional means are realized.

FIG. 2 is a flowchart showing a procedure of an object identifier extraction method executed by using the object identifier extraction device 100. Hereinafter, the object identifier extraction method according to the present embodiment will be described with reference to FIGS. 1 and 2.

First, the imaging unit 101 of the object identifier extraction device 100 acquires an image of the annular random pattern 111 formed on the object 110 (step S101).

Next, the extraction unit 102 of the object identification extraction device 100 extracts the feature amount from the image of the circular random pattern 111 acquired by the imaging unit 101 using the iris recognition algorithm (step S102). Next, the extraction unit 102 outputs the extracted feature amount as the object identifier 120 of the object 110 (step S103).

Subsequently, the outline of step S102 of FIG. 2, that is, the outline of the process of extracting the feature amount from the image of the circular random pattern 111 by using the iris authentication algorithm will be described. In the present embodiment, for example, the Daugman algorithm can be used as the iris authentication algorithm.

In the process of step S102, first, the image of the random pattern 111, which is the image of the portion corresponding to the iris, is extracted from the images (the image including the random pattern 111 and its surroundings) acquired by the imaging unit 101. Next, the extracted image of the random pattern 111 is subjected to a lossy compression process based on a predetermined mathematical method to generate a bit string including information representing the characteristics unique to the random pattern 111.

The bit string generated by performing mathematical processing on the image of the random pattern 111 corresponding to the iris in this way is obtained by extracting the features peculiar to the random pattern 111 and is formed on another object. It contains enough basic information to make a statistically meaningful comparison with the random pattern. The above bit string becomes the object identifier 120 in this embodiment.

For example, in the Daugman algorithm, a wavelet transform by a Gabor filter is used to extract a spatial frequency range having a predetermined SN ratio in consideration of performance such as the resolution of the imaging unit 101 from a random pattern 111 corresponding to an iris image. As a result, a complex number group including the local amplitude and phase information of the random pattern 111 is acquired. Then, based on the complex number group, a bit string of 2048 bits including only the sign bit of the complex number represented by the Gabor region, which does not include the amplitude information, is acquired as a bit string corresponding to the random pattern 111. Since the Daugman algorithm is widely used in general iris recognition, a more detailed description of the algorithm will be omitted.

As described above, according to the present embodiment, the technology cultivated in iris recognition can be applied to the certification of objects such as industrial products and commodities. That is, the object identifier can be extracted using a well-known algorithm that extracts and encodes features from the iris image. The reason is that since the shape of the random pattern 111 processed on the object 110 is similar to that of the iris, the feature quantity from the random pattern 111 is used by using a well-known algorithm that extracts and encodes the features from the iris image. This is because can be extracted.
Generally, in biometric authentication technology, various technical elements have been developed in order to recognize a large number of individuals with high accuracy. Therefore, according to the present embodiment, it is possible to realize a low-cost and high-performance object identifier extraction device by applying these techniques to industrial products and commodities.
Further, in order to extract the object identifier by the technique described in Patent Documents 1 to 3, a rectangle or a rectangle is placed on the identification target so that the size, orientation, and position of the above "predetermined area" can be accurately recognized. It is necessary to form an accurate pattern such as a logo. For this reason, accurate printing technology and advanced processing means without blurring or chipping are required for pattern formation, and there is a problem that the cost and processing time for extracting the identifier are high and the productivity is lowered. is there. In particular, in order to use it for identifying mass-produced industrial products / commodities, it is necessary to extract identifiers at low cost and at high speed, and there is a problem that it is often not applicable. According to this embodiment, such a problem can be solved.
Further, according to the present embodiment, it is possible to authenticate a large amount of objects such as industrial products and commodities by an extremely simple and low-cost means such as applying a drop of ink in order to impart a circular random pattern. be able to. When such a simple random pattern is used, it is difficult to apply the techniques described in Patent Documents 1 to 3, which are premised on the accurate formation of the pattern.
Further, in the present embodiment, by making the size of the circular random pattern 111 about the same as the size of the human iris, the object is authenticated using the imaging means (camera) used in the human iris recognition system. It can be performed. In addition, various existing technologies for improving the authentication accuracy by using the image pickup means can be diverted to the object authentication.
Instead of making the size of the circular random pattern 111 itself about the same as the size of the human iris, the image of the circular random pattern 111 is made about the same size as the human iris existing at the same position as the pattern. The conversion lens 1011 for magnification correction to be converted may be attached to the photographing unit 101. By using such a conversion lens 1101, it is possible to obtain the same effect as described above when the size of the annular random pattern 111 itself is made to be about the same as the size of the human iris. A similar conversion lens for magnification correction may be used in an embodiment described later in which a random pattern having an indefinite shape or a random pattern having a circular shape is used.

[Second Embodiment]
Referring to FIG. 3, the object identifier extraction device 200 according to the second embodiment of the present invention has a function of extracting the object identifier 220 of the object 210.

The object 210 is a target for extracting an object identifier, and is similar to the object 110 in FIG. An irregularly shaped random pattern 211 is processed on any surface of the object 210. In this example, the random pattern 211 has an indefinite shape, but it may have a predetermined shape such as a rectangle. The random pattern 211 is processed into a dot-shaped region 212 in which a part of the region is not a random pattern. Here, it is desirable that the surface of the object 210 on which the random pattern 211 is processed has a color, material, pattern, or the like that is visually distinguishable from the random pattern 211. Further, it is desirable that the dot-shaped region 212 is generated near the center of the random pattern 211. Further, it is desirable that the dot-shaped region 212 has a color, material, pattern, or the like that is visually distinguishable from the random pattern 211. Further, the size of the irregularly shaped random pattern 211 may be about the same as or slightly larger than the size of the human iris. However, the size is not limited to such a size, and the size of the random pattern 211 may be smaller than the size of the human iris or may be sufficiently large. The same applies to the embodiment described later, which uses a random pattern having an indefinite shape.

The method of processing the random pattern 211 on the surface of the object 210 may be the same as the method of processing the random pattern 111 on the surface of the object 110 of FIG. Further, the method of processing the dot-shaped region 212 is arbitrary. For example, in the dot-shaped region 212, ordinary printing ink, paint, or the like that does not contain fine particles is applied to the surface of the object 210 on which the random pattern 211 is processed by using a pen, a stamp, or the like. It may be produced by solidifying. Alternatively, when the random pattern 211 is processed on the surface of the object 210 by a laser processing machine, the original surface of the object 210 is left as it is by preventing the laser beam from shining on a part of the region. The region 212 may be generated. Alternatively, when the random pattern 211 on the surface originally possessed by the object 210 is used, the point-shaped region 212 may be processed on the random pattern 211. For example, for a product with a satin finish on the surface, the satin finish on the surface that naturally occurs in the manufacturing process of the product may be a random pattern, and the random pattern is not processed, but simply as a reference point. You only have to hit one point in the area 212 of.

The object identifier extraction device 200 has an imaging unit 201 and an extraction unit 202 as main functional units. The imaging unit 201 has the same function as the imaging unit 101 of FIG.

The extraction unit 202 has a function of extracting the feature amount of the random pattern 211 from the image acquired by the image pickup unit 201 and outputting the extracted feature amount as the object identifier 220 of the object 210. The extraction unit 202 uses an iris recognition algorithm for extracting the feature amount.

As shown in FIG. 18, for example, the object identifier extraction device 200 includes an information processing device 740 having an arithmetic processing unit 741 such as one or more microprocessors, a storage unit 742 such as a memory or a hard disk, and a camera 743, and a program. It can be realized with 744. The information processing device 740 may be, for example, a smartphone. The program 744 is read into the memory from an external computer-readable recording medium at the time of starting up the information processing device 740, and controls the operation of the arithmetic processing unit 741 so that the imaging unit 201 is placed on the arithmetic processing unit 741. , The extraction unit 202, and other functional means are realized.

FIG. 4 is a flowchart showing the procedure of the object identifier extraction method executed by using the object identifier extraction device 200. Hereinafter, the object identifier extraction method according to the present embodiment will be described with reference to FIGS. 3 and 4.

First, the imaging unit 201 of the object identifier extraction device 200 acquires an image of an irregularly shaped random pattern 211 formed on the object 210 (step S201).

Next, the extraction unit 202 of the object identification extraction device 200 cuts out an image of the inside of a circle of a predetermined size centered on the center of the point-shaped region 212 from the image of the random pattern 211 (step S202). For example, as shown in FIG. 5, a circle 214 having a radius r centered on the center 213 of the point-shaped region 212 formed on the surface of the object 210 is drawn, and an image region inside the circle 214 is cut out. The circle 214 may be a circle having a predetermined fixed length as a radius r. Alternatively, the circle 214 may be a circle having an area of a predetermined ratio (for example, about several percent) of the area of the random pattern 211. The image inside the circle cut out in this way becomes a circular random pattern.

Next, the extraction unit 202 extracts a feature amount from the image region of the circular random pattern cut out by using the iris recognition algorithm (step S203). In this step S203, the process of extracting the feature amount from the annular image region using the iris authentication algorithm is the same as in step S102 of FIG. Next, the extraction unit 202 outputs the extracted feature amount as the object identifier 220 of the object 210 (step S204).

As described above, according to the present embodiment, the object identifier 220 is extracted from the irregularly shaped random pattern 211 processed on the object 210 by using a well-known algorithm that extracts and encodes features from the iris image. Can be done. The reason is that a part of the area of the random pattern is processed into a point-shaped area 212 that is not a random pattern, and the extraction unit 202 sets a predetermined length from the center of the point-shaped area as a radius in the image of the random pattern. This is because the image in the circle is cut out as a circular image area and the iris recognition algorithm is applied.

[Third Embodiment]
Referring to FIG. 6, the object identifier extraction device 300 according to the third embodiment of the present invention has a function of extracting the object identifier 320 of the object 310.

The object 310 is a target for extracting an object identifier, and is similar to the object 110 in FIG. A circular random pattern 311 is processed on any surface of the object 310. Here, it is desirable that the surface of the object 310 on which the random pattern 311 is processed has a color, material, pattern, or the like that is visually distinguishable from the random pattern 311. The method of processing the random pattern 311 on the surface of the object 310 may be the same as the method of processing the random pattern 111 on the surface of the object 110 of FIG. Further, the size of the random pattern 311 may be about the same as the size of the human iris. However, the size is not limited to such a size, and the size of the random pattern 311 may be smaller or larger than the size of the human iris. The same applies to the embodiment described later, which uses a circular random pattern.

The object identifier extraction device 300 has an imaging unit 301 and an extraction unit 302 as main functional units. The imaging unit 301 has the same function as the imaging unit 101 of FIG.

The extraction unit 302 has a function of extracting the feature amount of the random pattern 311 from the image acquired by the image pickup unit 301 and outputting the extracted feature amount as the object identifier 320 of the object 310. The extraction unit 302 uses an iris recognition algorithm for extracting the feature amount.

As shown in FIG. 18, for example, the object identifier extraction device 300 includes an information processing device 740 having an arithmetic processing unit 741 such as one or more microprocessors, a storage unit 742 such as a memory or a hard disk, and a camera 743, and a program. It can be realized with 744. The information processing device 740 may be, for example, a smartphone. The program 744 is read into the memory from an external computer-readable recording medium at the time of starting up the information processing device 740, and controls the operation of the arithmetic processing unit 741 so that the imaging unit 301 is placed on the arithmetic processing unit 741. , Realizes functional means such as extraction unit 302.

FIG. 7 is a flowchart showing the procedure of the object identifier extraction method executed by using the object identifier extraction device 300. Hereinafter, the object identifier extraction method according to the present embodiment will be described with reference to FIGS. 6 and 7.

First, the imaging unit 301 of the object identifier extraction device 300 acquires an image of the circular random pattern 311 formed on the object 310 (step S301).

Next, as shown in FIG. 8, for example, the extraction unit 302 of the object identification extraction device 300 generates a circular image 312 of a predetermined size centered on the center of gravity of the image of the random pattern 311 on the image of the random pattern 311. (Step S302). The circular image 312 may be a circular image having a predetermined fixed value as the radius. Alternatively, the circular image 312 may be a circular image having an area of a predetermined ratio (for example, about 20 to 40%) of the area of the random pattern 311. The circular image 312 is an image filled with one color such as black or blue, and is not a random pattern. By performing such image processing, the random pattern 311 becomes circular.

Next, the extraction unit 302 extracts a feature amount from the image region of the circular random pattern 311 using the iris recognition algorithm (step S303). In step S303, the process of extracting the feature amount from the annular image region using the iris recognition algorithm is the same as in step S102 of FIG. Next, the extraction unit 302 outputs the extracted feature amount as the object identifier 320 of the object 310 (step S304).

Thus, according to the present embodiment, the object identifier 320 can be extracted from the circular random pattern 311 machined on the object 310 using a well-known algorithm that extracts and encodes features from the iris image. it can. The reason is that the extraction unit 302 generates a circular random pattern by generating a smaller circular image 312 on the image of the circular random pattern 311 and applies the iris recognition algorithm.

[Fourth Embodiment]
Referring to FIG. 9, the object identifier extraction device 400 according to the fourth embodiment of the present invention has a function of extracting the object identifier 420 of the object 410.

The object 410 is a target for extracting an object identifier, and is similar to the object 110 in FIG. An irregularly shaped random pattern 411 is processed on any surface of the object 410. In this example, the random pattern 411 has an indefinite shape, but it may have a predetermined shape such as a rectangle. Further, it is desirable that the surface of the object 410 on which the random pattern 411 is processed has a color, material, pattern or the like that can be visually distinguished from the random pattern 411. The method of processing the random pattern 411 on the surface of the object 410 may be the same as the method of processing the random pattern 111 on the surface of the object 110 of FIG.

The object identifier extraction device 400 has an imaging unit 401 and an extraction unit 402 as main functional units. The image pickup unit 401 has the same function as the image pickup unit 101 of FIG.

The extraction unit 402 has a function of extracting the feature amount of the random pattern 411 from the image acquired by the image pickup unit 401 and outputting the extracted feature amount as the object identifier 420 of the object 410. The extraction unit 402 uses an iris recognition algorithm for extracting the feature amount.

As shown in FIG. 18, for example, the object identifier extraction device 400 includes an information processing device 740 having an arithmetic processing unit 741 such as one or more microprocessors, a storage unit 742 such as a memory or a hard disk, and a camera 743, and a program. It can be realized with 744. The information processing device 740 may be, for example, a smartphone. The program 744 is read into the memory from an external computer-readable recording medium at the time of starting up the information processing device 740, and controls the operation of the arithmetic processing unit 741 so that the imaging unit 401 is placed on the arithmetic processing unit 741. , Realize functional means such as extraction unit 402.

FIG. 10 is a flowchart showing the procedure of the object identifier extraction method executed by using the object identifier extraction device 400. Hereinafter, the object identifier extraction method according to the present embodiment will be described with reference to FIGS. 9 and 10.

First, the imaging unit 401 of the object identifier extraction device 400 acquires an image of an irregularly shaped random pattern 411 formed on the object 410 (step S401).

Next, the extraction unit 402 of the object identification extraction device 400 generates a circular image 412 centered on the center of gravity 413 of the image of the random pattern 411 on the image of the random pattern 411, for example, as shown in FIG. Step S402). FIG. 11 is similar to FIG. 5, but the circular image 212 of FIG. 5 is processed on the surface of the object, whereas the circular image 412 of FIG. 11 is on the image of the object by image processing. The two differ in that they are virtually generated in. The circular image 412 may be a circular image having a predetermined fixed length as a radius. Alternatively, the circular image 412 may be a circular image having an area of a predetermined ratio (for example, about 20 to 40%) of the area of the random pattern 411. The circular image 412 is an image filled with one color such as black or blue, and is not a random pattern.

Next, as shown in FIG. 11, the extraction unit 402 draws an image inside a circle 414 having a size larger than that of the circular image 412, centered on the center of gravity of the image of the random pattern 411, among the images of the random pattern 411. Cut out (step S403). This cut-out image becomes an image area of a circular random pattern. The circle 414 may be an image of a circle having a predetermined fixed length as a radius. Alternatively, the circle 414 may be a circle having an area of a predetermined ratio (for example, about 50 to 70%) of the area of the random pattern 411.

Next, the extraction unit 402 extracts a feature amount from the image region of the circular random pattern cut out by using the iris recognition algorithm (step S404). In step S404, the process of extracting the feature amount from the annular image region using the iris recognition algorithm is the same as in step S102 of FIG. Next, the extraction unit 402 outputs the extracted feature amount as the object identifier 420 of the object 410 (step S405).

Thus, according to the present embodiment, the object identifier 420 is extracted from the irregularly shaped random pattern 411 processed on the object 410 using a well-known algorithm that extracts and encodes features from the iris image. Can be done. The reason is that the extraction unit 402 generates an image 412 of points centered on the center of gravity of the image of the random pattern 411 on the image 411 of the random pattern, and further, an image of the random pattern 411 among the images 411 of the random pattern. This is to generate an annular random pattern and apply the iris authentication algorithm by cutting out an image inside a circle of a predetermined size centered on the center of gravity of.

[Fifth Embodiment]
Referring to FIG. 12, the object identifier registration device 500 according to the fifth embodiment of the present invention has a function of processing a random pattern 511 on the surface of the object 510, and the processed random pattern 511 to the object identifier 520 of the object 510. It has a function to extract.

The object 510 is a target for extracting an object identifier by processing a random pattern 511. The object 510 is, for example, an industrial product or a product package. The object 510 may be an individual's belongings (business card, notebook, etc.) or a part of the body (for example, a finger) excluding the individual's iris.

The object identifier registration device 500 has an imaging unit 501, an extraction unit 502, and a processing unit 503 as main functional units. Of these, the imaging unit 501 and the extraction unit 502 are the imaging units 101, 201, 301, 401 and the extraction units 102, 202 of the object identifier extraction devices 100, 200, 300, 400 in the first to fourth embodiments. It has the same function as 302 and 402.

The processing unit 503 has a function of processing a random pattern 511 on the surface of the object 510. The method in which the processing unit 503 processes the random pattern 511 on the surface of the object 510 is arbitrary. For example, the processing unit 503 may generate a random pattern 511 by applying a printing ink, a paint, or the like mixed with fine particles to the surface of the object 510 using a pen, a stamp, or the like and solidifying the surface. As the fine particles, fine particles such as metal powder and glass powder, taggants described in Patent Documents 1 to 3 and the like can be used. It is desirable that the fine grains are grains having different reflection characteristics from the material (excluding the fine grains) constituting the random pattern 511. Further, it is desirable that the fine grains are non-uniformly contained in the random pattern 511. Alternatively, the processing unit 503 may generate a random pattern 511 by processing a random pattern on the surface of the object 510 with a laser processing machine. Alternatively, when the surface of the object 510 is satin finished, the processing unit 503 may generate a random pattern 511 using the satin finish.

Here, when the extraction unit 502 has the same function as the extraction unit 102 in the first embodiment, the processing unit 503 processes an annular random pattern 111 as shown in FIG. 1 on the surface of the object 510. Further, in the processing unit 503, when the extraction unit 502 has the same function as the extraction unit 202 in the second embodiment, a part of the region as shown in FIG. 3 is a random region that is not a random pattern. The pattern 211 is processed into the surface of the object 510. Further, when the extraction unit 502 has the same function as the extraction unit 302 in the third embodiment, the processing unit 503 processes a substantially circular random pattern 311 as shown in FIG. 6 on the surface of the object 510. Further, in the processing unit 503, when the extraction unit 502 has the same function as the extraction unit 402 in the fourth embodiment, a part of the region as shown in FIG. 9 is a random region that is not a random pattern. The pattern 411 is processed into the surface of the object 510.

As shown in FIG. 18, for example, the object identifier registration device 500 includes an information processing device 740 having an arithmetic processing unit 741 such as one or more microprocessors, a storage unit 742 such as a memory or a hard disk, and a camera 743, and a program. It can be realized with 744. The information processing device 740 may be, for example, a smartphone. The program 744 is read into the memory from an external computer-readable recording medium at the time of starting up the information processing device 740, and controls the operation of the arithmetic processing unit 741 to control the operation of the arithmetic processing unit 741 on the imaging unit 501 on the arithmetic processing unit 741. , The extraction unit 502 and the like are realized.

FIG. 13 is a flowchart showing the procedure of the object identifier registration method executed by using the object identifier registration device 500. Hereinafter, the object identifier registration method according to the present embodiment will be described with reference to FIGS. 12 and 13.

First, the processing unit 503 of the object identifier registration device 500 processes a random pattern 511 on the surface of the object 510 (step S501).

Next, the imaging unit 501 acquires an image of the random pattern 511 processed on the object 510 (step S502).

Next, the extraction unit 502 extracts a feature amount from the image of the random pattern 511 acquired by the imaging unit 501 by using the iris authentication algorithm (step S503). This step S503 is performed in the same manner as steps S102 in FIG. 2, steps S202 to S203 in FIG. 4, steps S302 to S303 in FIG. 7, and steps S402 to S404 in FIG.

Finally, the extraction unit 502 outputs the extracted feature amount as the object identifier 520 of the object 510 (step S504).

As described above, according to the present embodiment, after processing the random pattern 511 on the surface of the object 510, the object is processed from the processed random pattern 511 using a well-known algorithm that extracts and encodes features from the iris image. The identifier can be extracted.

[Sixth Embodiment]
Referring to FIG. 14, the object identifier registration device 600 according to the sixth embodiment of the present invention has a function of processing a random pattern 611 on the surface of the object 610 and extracts an object identifier of the object 610 from the processed random pattern 611. It has a function of registering data including an extracted object identifier in the storage device 620. Here, the object 610 and the random pattern 611 are the same as the object 510 and the random pattern 511 shown in FIG. 12 in the fifth embodiment of the present invention.

The object identifier registration device 600 has an imaging unit 601, an extraction unit 602, a processing unit 603, and a registration unit 604 as main functional units. Of these, the imaging unit 601, the extraction unit 602, and the processing unit 603 have the same functions as the imaging unit 501, the extraction unit 502, and the processing unit 503 in the fifth embodiment.

The registration unit 604 has a function of registering the object identifier of the object 610 extracted by the extraction unit 502 in the storage device 620 as the identification information of the registered object. The storage device 620 may be, for example, a storage of a server device accessible from the object identifier registration device 600 via a network. Alternatively, the storage device 620 may be a local storage device connected to the object identification / registration device 600.

As shown in FIG. 18, for example, the object identifier registration device 600 includes an information processing device 740 having an arithmetic processing unit 741 such as one or more microprocessors, a storage unit 742 such as a memory or a hard disk, and a camera 743, and a program. It can be realized with 744. The information processing device 740 may be, for example, a smartphone. The program 744 is read into the memory from an external computer-readable recording medium at the time of starting up the information processing device 740, and controls the operation of the arithmetic processing unit 741 to control the operation of the arithmetic processing unit 741 on the imaging unit 601 on the arithmetic processing unit 741. , Extraction unit 602, registration unit 604, and other functional means are realized.

FIG. 15 is a flowchart showing a procedure of an object identifier registration method executed by using the object identifier registration device 600. Hereinafter, the object identifier registration method according to the present embodiment will be described with reference to FIGS. 14 and 15.

First, the processing unit 603 of the object identifier registration device 600 processes a random pattern 611 on the surface of the object 610 (step S601). Next, the imaging unit 601 acquires an image of the random pattern 611 processed on the object 610 (step S602). Next, the extraction unit 602 extracts a feature amount from the image of the random pattern 611 acquired by the imaging unit 601 using the iris authentication algorithm (step S603). These steps S601 to S603 are performed in the same manner as steps S501 to S503 in the fifth embodiment.

Finally, the registration unit 604 registers the data including the feature amount extracted by the extraction unit 602 as the object identifier 520 of the object 510 in the storage device 620 (step S604).

FIG. 16 shows an example of data registered in the storage device 620. In this example, the storage device 620 stores data having an object identifier of the object and one or more attribute values of the object. For example, when the object is an industrial product or a product package, the model number, the production lot number, the number of the machine or tool used to process the object, and the like are examples of attribute values. The object identifier is a bit string having a number of bits N (≧ 2). When using Daugman's algorithm as the iris recognition algorithm, the object identifier is a 2048-bit bit string.

As described above, according to the present embodiment, after processing the random pattern 611 on the surface of the object 610, the object is processed from the processed random pattern 511 using a well-known algorithm that extracts and encodes features from the iris image. The identifier can be extracted and registered in the storage device 620 as the identification information of the registered object.

[7th Embodiment]
Referring to FIG. 17, the identification collation device 700 according to the seventh embodiment of the present invention has a function of identifying and collating the object 710.

The object 710 is an object to be identified / collated, such as an industrial product or a product package. Further, the object 710 may be an individual's belongings (business card, notebook, etc.) or a part of the body (for example, a finger) excluding the individual's iris.

The identification collation device 700 has an imaging unit 701, an extraction unit 702, and a determination unit 703. Of these, the imaging unit 701 and the extraction unit 702 have the same functions as the imaging unit 501 and the extraction unit 502 in the fifth embodiment shown in FIG.

The determination unit 703 compares the object identifier of the object 710 extracted by the extraction unit 702 with the object identifier of the registered object stored in the storage device 720, and identifies and collates the object based on the comparison result. It has a function to make a judgment. The storage device 720 is, for example, the same as the storage device 620 in the sixth embodiment, and data having an individual identifier of a registered object and its attribute value is stored in advance.

As shown in FIG. 18, for example, the identification collation device 700 includes an information processing device 740 having an arithmetic processing unit 741 such as one or more microprocessors, a storage unit 742 such as a memory or a hard disk, and a camera 743, and a program 744. It can be realized with. The information processing device 740 may be, for example, a smartphone. The program 744 is read into the memory from an external computer-readable recording medium at the time of starting up the information processing device 740, and controls the operation of the arithmetic processing unit 741 to control the operation of the arithmetic processing unit 741 on the imaging unit 701 on the arithmetic processing unit 741. , Extraction unit 702, determination unit 703, and other functional means are realized.

FIG. 19 is a flowchart showing the procedure of the identification collation method executed by using the identification collation device 700. Hereinafter, the identification collation method according to the present embodiment will be described with reference to FIGS. 17 and 19.

First, the imaging unit 701 of the identification collation device 700 acquires an image of the random pattern 711 processed on the object 710 (step S701). Next, the extraction unit 702 extracts a feature amount from the image of the random pattern 711 acquired by the imaging unit 701 using the iris recognition algorithm (step S702).

Next, the determination unit 703 of the identification collation device 700 compares the object identifier of the extracted object 710 with the object identifier of the registered object stored in the storage unit 720, and based on the comparison result, the object. Is identified / determined (step S703). For example, when using the Daugman algorithm, the determination unit 703 calculates the Hamming distance between the bit string of the object identifier of the object 710 and the object identifier of the registered object, and if the Hamming distance is equal to or less than the threshold value, the two objects. It is determined that the identifiers are the same, otherwise it is determined that they are not the same. The determination unit 703 finds the object identifier of the same registered object as the object identifier of the object 710, or finishes the comparison with the object identifiers of all the registered objects, whichever is earlier. Is a condition, the repetition of the object identification / judgment process is completed. Then, the determination unit 703 outputs the determination result 730 (step S704). The determination result 730 may indicate whether or not the identification / collation is successful. Further, in the case of success, the determination result 730 may have the attribute value of the object stored in the storage unit 720 corresponding to the individual identifier of the registered object determined to be the same.

As described above, according to the present embodiment, the object identifier is extracted from the random pattern 711 processed on the surface of the object 710 by using the iris recognition algorithm, and the object is identified by comparing with the registered object identification information.・ Collation can be performed.

[8th Embodiment]
Referring to FIG. 20, the object identifier management system 800 according to the eighth embodiment of the present invention has a function of managing an object identifier for identifying / collating an object 810.

The object 810 is an object to be managed by assigning an object identifier such as an industrial product or a product package. In FIG. 20, only one object 810 to be managed by assigning an object identifier is shown, but in general, a large number of objects 810 are to be managed.

The object identifier management system 800 is composed of an object identifier registration device 801 and an identification collation device 802.

The object identifier registration device 801 has a function of processing a random pattern 811 on the surface of the object 810, a function of capturing an image of the object 810 processed by the random pattern 811, and a random pattern 811 among the images of the captured object 810. It has a function of extracting the features of the including annular image region as identification information of the object 810 and a function of registering the identification information of the extracted object 810 in the storage device 820 as the identification information of the registered object. The object identifier registration device 801 is composed of, for example, the object identifier registration device 600 according to the sixth embodiment of the present invention.

Further, the identification collation device 802 identifies the object 810 with a function of imaging an object 810 having a random pattern 811 processed on the surface thereof and a feature of an annular image region including the random pattern 811 among the images of the captured object 810. The function of extracting as information is compared with the identification information of the extracted object 810 and the identification information of the registered object stored in the storage device 820, and the identification / collation of the object is determined based on the comparison result. Has a function. The identification collation device 802 is composed of, for example, the identification collation device 700 according to the seventh embodiment of the present invention.

FIG. 21 is a flowchart showing a procedure of an object identifier management method executed by using the object identifier management system 800. Hereinafter, the object identifier management method according to the present embodiment will be described with reference to FIGS. 20 and 21.

First, the object identifier registration device 801 processes a random pattern 811 on the surface of the object 810 (step S801). Next, the object identifier registration device 801 acquires an image of the random pattern 811 processed on the object 810 (step S802). Next, the object identifier registration device 801 extracts a feature amount from the acquired image of the random pattern 811 by using the iris recognition algorithm (step S803). Next, the object identifier registration device 801 registers the data including the extracted feature amount as the object identifier of the object 810 in the storage device 820 (step S804). These steps S801 to S804 are performed in the same manner as steps S601 to S604 in the sixth embodiment.

On the other hand, the identification collation device 802 acquires an image of the random pattern 811 processed on the object 810 (step S805). Next, the identification collation device 802 extracts the feature amount from the acquired image of the random pattern 811 by using the iris authentication algorithm (step S806). Next, the identification collation device 802 compares the object identifier of the extracted object 810 with the object identifier of the registered object stored in the storage device 820, and identifies / determines the object based on the comparison result. (Step S807). Next, the identification collation device 802 outputs the determination result 830 (step S808). These steps S805 to S808 are performed in the same manner as steps S701 to S704 in the seventh embodiment.

In the flowchart of FIG. 21, following the processing of steps S801 to S804 by the object identifier registration device 801, the processing of steps S805 to S808 by the identification collation device 802 was executed. However, the procedure is not limited to such a procedure, and the processing of steps S801 to S804 by the object identifier registration device 801 may be repeated a plurality of times for different objects 810. Further, the processing of steps S805 to S808 by the identification collation device 802 may be repeated a plurality of times for different objects 810. Alternatively, the processing of steps S801 to S804 by the object identifier registration device 801 and the processing of steps S805 to S808 by the identification collation device 802 may be performed on different objects 810 in parallel.

As described above, according to the present embodiment, after processing the random pattern 811 on the surface of the object 810, the object is processed from the processed random pattern 811 using a well-known algorithm that extracts and encodes features from the iris image. The identifier can be extracted and registered in the storage device 820 as the identification information of the registered object. Further, according to the present embodiment, the object identifier is extracted from the random pattern 811 processed on the surface of the object 810 by using the iris recognition algorithm, and the object is identified by comparing with the registered object identification information. Collation can be performed.

[9th Embodiment]
In this embodiment, an embodiment in which the object identifier registration device 600 according to the sixth embodiment is more embodied will be described.

Referring to FIG. 22, the object identifier registration device 900 according to the ninth embodiment of the present invention has a structure in which the ink pen 901 and the pattern imaging unit 902 are housed in the housing 903. The ink pen 901 corresponds to the processing unit 603, and the pattern imaging unit 902 corresponds to the imaging unit 601, the extraction unit 602, and the registration unit 604. Although not shown, the housing 903 includes a pattern imaging unit 902, limit switches 909 and 910 described later, and a battery for operating the illuminator 914. The housing 903 is configured as a portable device operated by a built-in battery.

The housing 903 is composed of a cylindrical container 904 and a lid 905 that is loosely fitted to the container 904. Both the upper surface and the lower surface of the container 904 are open, and a light window 906 and an illuminator 914 are provided at the lower part of the side surface. A pattern imaging unit 902 is fixed inside the container 904, and a movable mirror 908, limit switches 909, and 910 are further attached. The on / off state of the limit switches 909 and 910 is transmitted to the pattern imaging unit 902 by a signal line (not shown). The illuminator 914 is composed of a white lamp, a flash lamp, or the like, and its lighting and extinguishing are controlled by a signal transmitted from the pattern imaging unit 902 through a signal line (not shown).

The movable mirror 908 is rotatably attached to a position above the bottom surface of the container 904 by a predetermined length around a shaft 908a extending in the direction perpendicular to the paper surface. Further, the movable mirror 908 is urged counterclockwise by the tension spring 908b, and the movable mirror 908 is stopped in a state where the end portion of the mirror is in contact with the limit switch 909. In this stopped state, the angle formed by the mirror surface of the movable mirror 908 and the bottom surface of the container 904 is approximately 45 degrees. The pattern imaging unit 902 is fixed to the container 904 so that the imaging field of view of the built-in camera can see the center of the mirror surface of the movable mirror 908 from the side.

On the other hand, the lid 905 is attached to the container 904 in the vertical direction via an elastic body 911 such as a spring or a sponge. A fixing plate 913 is attached to the inside of the lid 905 via a connecting member 912, and an ink pen 901 is attached to the lower surface of the fixing plate 913 directly above the movable mirror 908.

The pen shaft of the ink pen 901 extends directly below. When the lid 905 is not pressed from above, the pen tip of the ink pen 901 is in a position (standby position) that does not come into contact with the movable mirror 908, as shown in FIG. On the other hand, when the lid 905 is pressed from above, as shown in FIG. 23, the pen tip of the ink pen 910 that moves up and down integrally with the lid 905 pushes down one end of the movable mirror 908 and reaches the lower surface of the container 904. It is in the position (generation position). When the movable mirror 908 is pushed down by the pen tip, the end portion of the movable mirror 908 is separated from the limit switch 909, and the limit switch 909 is turned off. When the pen tip of the ink pen 901 reaches the lower surface of the container 904, the fixing plate 913 comes into contact with the limit switch 910 and the limit switch 910 is turned on, as shown in FIG. 23.

The ink pen 901 is a pen that uses ink mixed with fine particles. As the fine particles, fine particles such as metal powder and glass powder, taggants described in Patent Documents 1 to 3 and the like can be used. It is desirable that the fine particles are particles having different reflection characteristics from the material (excluding the fine particles) constituting the ink. Further, it is desirable that the fine particles are non-uniformly contained in the ink. As the ink pen 901, for example, a pen commercially available under the names of a lame pen, a pen with lame, a highlighter with lame, and the like can be used. Of course, a pen specially designed for the present invention may be used, or the ink of a commercially available pen may be replaced with an ink mixed with fine particles.

Next, the operation of the object identifier registration device 900 will be described with reference to FIGS. 22 and 23.

When the user uses the object identifier registration device 900 to generate a pattern for individual identification on an object, the user places the housing 903 on the object for which the pattern is to be generated so that the lower surface of the container 904 is in contact with the object. At this time, the object identifier registration device 900 is in a state as shown in FIG. 22, the limit switch 909 is in the on state, and the limit switch 910 is in the off state.

When the user pushes down the lid 905 of the housing 903 placed on the object, the state gradually shifts from the state shown in FIG. 22 to the state shown in FIG. 23. First, the lid 905 descends downward against the repulsive force of the elastic body 911, and one end of the movable mirror 908 is pushed down by the pen tip of the ink pen 901 linked to the movement of the lid 905, and the limit switch 909 is turned on. Changes to the off state. When the user further pushes the lid 905 to a position where it cannot be pushed down any further, the pen tip of the ink pen 901 built in the container 904 comes into contact with the surface of the object, and the ink flowing out from the pen tip causes a small dot on the object. Is drawn. As described above, since fine particles are unevenly mixed in the ink, the points to be drawn are fine particles (represented by black circles in the figure) as schematically shown in FIG. 24. Points (dots) contained at random positions. Further, when the lid 905 is lowered until the pen tip of the ink pen 901 comes into contact with the surface of the object, the limit switch 910 changes from the off state to the on state.

Next, when the user releases the force pushing down the lid 905, the repulsive force of the elastic body 911 causes the lid 905 to rise toward the original position. In the process, the pen tip of the ink pen 901 first rises away from the object, and the limit switch 910 changes from the on state to the off state. When the lid 905 is further raised to the state shown in FIG. 22, the movable mirror 908 returns to the original position by the force of the tension spring 908b, and the limit switch 909 changes from the on state to the off state.

The pattern imaging unit 902 detects the generation of the pattern based on the states of the limit switches 909 and 910, and detects the timing of imaging the generated pattern. For example, when the limit switch 909 changes from the on state to the off state and then the limit switch 910 changes from the off state to the on state, that is, when the pattern imaging unit 902 transitions from the state of FIG. 22 to the state of FIG. It is determined that the pattern is drawn on the object by the ink pen 901. Further, the pattern imaging unit 902 draws the pattern when the limit switch 909 changes from the off state to the on state after determining that the pattern has been drawn by the ink pen 901, that is, when the state of FIG. 23 changes to the state of FIG. 22. It is determined that it is time to capture the pattern, and the image of the drawn pattern is captured by the built-in camera. At this time, the pattern imaging unit 902 temporarily turns on the illuminator 914 as needed.

Next, the pattern imaging unit 902 will be described in detail.

FIG. 25 is a block diagram showing an example of the hardware configuration of the pattern imaging unit 902. The pattern imaging unit 902 of this example includes a camera 9021, an MPU (Micro Processor Unit) 9022, a ROM (Read Only Memory) 9023, a RAM (Random Access Memory) 9024, an interface circuit 9025, a communication circuit 9026, and the like. It consists of a bus 9027 that connects them to each other.

As the camera 9021, for example, a camera using a CCD (Charge Coupled Devices) image sensor or a CMOS (Completion Metal Oxide) image sensor can be used.

The ROM 9023 stores a program or the like executed by the MPU 9022. The RAM 9024 is used to store image data and the like captured by the camera 9021. The interface circuit 9025 is connected to the limit switches 909, 910 and the illuminator 914 through a signal line (not shown). The communication circuit 9026 has a function of wirelessly communicating with an external device.

FIG. 26 is a flowchart showing an example of the processing executed by the MPU 9022. Hereinafter, the processing executed by the MPU 9022 will be described with reference to FIGS. 22 to 26.

The MPU 9022 reads the state of the limit switch 909 through the interface circuit 9025 and monitors whether or not it is turned off (step S901). As described above, in the state of the object identifier registration device 900 as shown in FIG. 22, the limit switch 909 is in the on state, and the object identifier registration device 900 is placed on the object to generate a pattern, and the lid 905 is pushed downward. Then, one end of the movable mirror 908 is pushed down by the pen tip of the ink pen 901, and the limit switch 909 changes from the on state to the off state. When the MPU 9022 detects that the limit switch 909 has changed to the off state, it detects whether or not the limit switch 910 has changed to the on state (step S902).

When the limit switch 910 is turned on by pushing the lid 905 further to a position where the user cannot push it down any further, the MPU 9022 receives the pen tip of the ink pen 901 built in the container 904 in contact with the surface of the object. Then, it is determined that a pattern of small dots is drawn on the object by the ink flowing out from the pen tip, and then the limit switch 909 is monitored to change to the ON state (step S903). On the other hand, the MPU 9022 is drawn by the ink pen 901 when the limit switch 909 is returned to the on state after the limit switch 909 is turned off and before the limit switch 910 is turned on (YES in step S904). It is determined that the lid 905 has been returned to the original position without any problem, and the process returns to step S901.

Next, when the MPU 9022 detects that the limit switch 909 is turned on in step S903, it issues a command to the camera 9021 to take an image (step S905). The camera 9021 captures an image in response to this command. When the illuminance is insufficient, the camera 9021 temporarily turns on the illuminator 914 at the time of shooting through the interface circuit 9025. The state of the object identifier registration device 900 at the time of shooting is as shown in FIG. 22, and the camera 9021 can move an image of a pattern as shown in FIG. 24 drawn on the object with the ink pen 901. An image is taken as an image from directly above through the mirror 908, and the image data obtained by the image is stored in a predetermined area of the RAM 9024. In the present embodiment, the image of the pattern is captured at the timing when the ink pen 901 returns to the state shown in FIG. 22, but it is generated at the timing when the ink pen 901 moves upward by a certain distance from the state shown in FIG. The image of the pattern may be captured.

Next, the MPU 9022 analyzes the image data stored in the RAM 9024 and extracts the object identifier from the image of the drawn points (step S906). Specifically, the MPU 9022 has a circular shape centered on the center of gravity of the image of the random pattern on the image of the substantially circular random pattern as shown in FIG. 24, similarly to the third embodiment shown in FIG. By generating an image of, an annular image region is generated and an iris recognition algorithm is applied to extract an object identifier. Alternatively, the MPU 9022 generates an image of points centered on the center of gravity of the random pattern image as shown in FIG. 24 on the random pattern image, as in the fourth embodiment shown in FIG. An object identifier is extracted by cutting out an image in a circle having a predetermined length as a radius from the center of the above point among images of a random pattern as an annular image region and applying an iris recognition algorithm. Next, the MPU 9022 registers the data including the extracted object identifier in an external storage device (not shown) through the communication circuit 9026 (step S907). The MPU 9022 may include information stored in the ROM 9023 or the RAM 9024 in the transmission data, for example, information about the object that generated the pattern, identification information of the object identifier registration device 900, and attribute information such as the current time. Then, the MPU 9022 returns to the process of step S901 in preparation for the next imaging.

As described above, according to the present embodiment, the pattern imaging unit 902 detects that a pattern is generated on the object by the ink pen 901 and performs imaging, so that there is a discrepancy between the pattern generation and imaging. Can be prevented.

Further, according to the present embodiment, the pattern generated on the object can be captured as an image from directly above the object.

Further, according to the present embodiment, the object identifier can be extracted from the captured pattern and registered in an external storage device.

In the present embodiment, the movable mirror 908 is used, but the movable mirror 908 may be omitted. In this case, as shown in FIGS. 27 and 28, for example, the ink pen 901 is obliquely attached to the fixing plate 913 so that its tip is located above the center of the lower surface of the container 904. Further, the pattern imaging unit 902 is fixed to the container 904 so that the imaging direction of the built-in camera faces the center of the lower surface of the container 904. According to the configurations shown in FIGS. 27 and 28, the movable mirror 908 is omitted, so that the configuration is simplified. In addition, in FIG. 27 and FIG. 28, the limit switch 915 is a substitute for the limit switch 909, and is turned on when the lid 905 is not pressed.

If there is no problem even if a part of the ink pen 901 is captured in the image of the drawn pattern, the image is taken with the tip of the ink pen 901 in contact with the object (for example, the state of FIG. 23 or FIG. 28). It may be. Further, the pen tip of the ink pen 901 is configured to be able to move forward and backward in the axial direction so that the pen tip reaches the lower surface of the container 904 in the state of FIG. 22 or 27, and when the lid 905 is pushed down, the pen tip is pushed against the object. It may be configured to shrink in the applied state.

Further, in the present embodiment, the ink pen 901 is used, but instead of the ink pen 901, an ink dropping device that drops an appropriate amount (for example, one drop) of ink, an ink injector that sprays an appropriate amount of ink, and a brush or felt that sprays ink. You may use a coater which applies only an appropriate amount depending on the material. Alternatively, instead of the ink pen 901, a laser processing machine that forms a random pattern on an object may be used.

Further, in the present embodiment, each time a pattern is generated and the image is captured, the object identifier is extracted and registered in the storage device, but the captured image is temporarily stored in a memory (RAM or the like) and later. The object identifier may be extracted and registered in the storage device. Further, in the present embodiment, the object identifier is registered in the storage device, but instead of or in addition to the object identifier, the image of the captured pattern may be registered in the storage device.

[10th Embodiment]
In this embodiment, another embodiment in which the object identifier registration device 600 according to the sixth embodiment is more embodied will be described.

With reference to FIG. 29, in the individual identification device 1000 according to the tenth embodiment of the present invention, the ink pen 901 of the object identifier registration device 900 according to the ninth embodiment described with reference to FIG. 22 is the first. The pattern imaging unit 902 is replaced by the pattern imaging unit 1051 by being replaced by the inkjet nozzle 1031, the first inkjet 1032, the first ink cartridge 1033, the second inkjet nozzle 1041, the second inkjet 1042, and the second ink cartridge 1043. Has been replaced. In addition, the same members as those of the object identifier registration device 900 according to the ninth embodiment are designated by the same reference numerals.

The first inkjet nozzle 1031 and the second inkjet nozzle 1041 are arranged close to each other, and the nozzles extend directly downward toward the lower surface of the container 904. When the lid 905 is not pressed from above, the tips of the first inkjet nozzle 1031 and the second inkjet nozzle 1041 are in positions (standby positions) that do not come into contact with the movable mirror 908. On the other hand, in the state shown in FIG. 30 in which the lid 905 is pressed from above, the tips of the first inkjet nozzle 1031 and the second inkjet nozzle 1041 that move up and down integrally with the lid 905 push down one end of the movable mirror 908. , It is at a position (generation position) reaching near the lower surface of the container 904. When one end of the movable mirror 908 is pushed down by these nozzles, the end portion of the movable mirror 908 is separated from the limit switch 909, and the limit switch 909 is turned off. When the tip of the nozzle reaches the vicinity of the lower surface of the container 904, the fixing plate 913 comes into contact with the limit switch 910 and the limit switch 910 is turned on, as shown in FIG.

The first ink cartridge 1033 stores ink mixed with fine particles. As the fine particles, fine particles such as metal powder and glass powder, taggants described in Patent Documents 1 and 2 and the like can be used. It is desirable that the fine particles are particles having different reflection characteristics from the material (excluding the fine particles) constituting the ink. Further, it is desirable that the fine particles are non-uniformly contained in the ink. When the first inkjet 1032 receives an output command from the pattern imaging unit 1051 through a signal line (not shown), the tip of the first inkjet nozzle 1031 receives the ink supplied from the first ink cartridge 1033 at a predetermined pressure for a certain period of time. Ink out from.

On the other hand, the second ink cartridge 1043 stores ink that is not mixed with fine particles as described above. For example, the second ink cartridge 1043 stores ordinary ink of a single color such as black. When the second inkjet 1042 receives an output command from the pattern imaging unit 1051 through a signal line (not shown), the tip of the second inkjet nozzle 1041 receives the ink supplied from the second ink cartridge 1043 at a predetermined pressure for a certain period of time. Ink out from.

Here, the size of the points drawn by the first inkjet nozzle 1031 is set to be sufficiently larger than the size of the points drawn by the second inkjet nozzle 1041. For example, the first inkjet 1032 and the second inkjet 1042 have an ink outflow amount so that the amount of ink ejected per unit time is larger in the first inkjet nozzle 1031 than in the second inkjet nozzle 1041. Is configured to adjust. Further, the nozzle diameter of the first inkjet nozzle 1031 may be larger than that of the second inkjet nozzle 1041.

FIG. 31 schematically shows a pattern obtained by drawing a point on an object by the first inkjet nozzle 1031 and immediately drawing a small point on the point by the second inkjet nozzle 1041. .. In FIG. 31, the area 1061 corresponds to the point drawn by the first inkjet nozzle 1031 and the area 1062 corresponds to the point drawn by the second inkjet nozzle 1041. Region 1061 contains minute grains (represented by black circles in the figure) at random positions, but region 1062 does not. Therefore, the shape of the region 1061 containing fine grains is annular.

Next, the operation of the object identifier registration device 1000 will be described with reference to FIGS. 29 to 31.

When the user uses the object identifier registration device 1000 to generate a pattern for individual identification on an object, the user places the housing 903 on the object for which the pattern is to be generated so that the lower surface of the container 904 is in contact with the object. At this time, the object identifier registration device 1000 is in a state as shown in FIG. 29, the limit switch 909 is in the on state, and the limit switch 910 is in the off state.

When the user pushes down the lid 905 of the housing 903 placed on the object, the state gradually shifts from the state shown in FIG. 29 to the state shown in FIG. First, the lid 905 is lowered against the repulsive force of the elastic body 911, and one end of the movable mirror 908 is pushed down by the first inkjet nozzle 1031 and the second inkjet nozzle 1041 linked to the movement of the lid 905. , The limit switch 909 changes from the on state to the off state. When the user further pushes down the lid 905 to a position where it cannot be pushed down any further, the fixing plate 913 comes into contact with the limit switch 910, and the limit switch 910 changes from the off state to the on state. At this point, the tips of the first inkjet nozzle 1031 and the second inkjet nozzle 1041 reach the vicinity of the lower surface of the container 904. Therefore, by ejecting ink from the tips of the first inkjet nozzle 1031 and the second inkjet nozzle 1041, a pattern as shown in FIG. 31 can be drawn on the object.

Next, when the user releases the force pushing down the lid 905, the repulsive force of the elastic body 911 causes the lid 905 to rise toward the original position. In the process, the tips of the first inkjet nozzle 1031 and the second inkjet nozzle 1041 rise, and the limit switch 910 changes from the on state to the off state. When the lid 905 is further raised to reach the state shown in FIG. 29, the movable mirror 908 returns to the original position by the force of the tension spring 908b, and the limit switch 909 changes from the off state to the on state.

The pattern imaging unit 1051 detects the pattern generation timing and the timing of imaging the generated pattern based on the states of the limit switches 909 and 910. For example, when the limit switch 909 changes from the on state to the off state and then the limit switch 910 changes from the off state to the on state, that is, when the pattern imaging unit 1051 transitions from the state of FIG. 29 to the state of FIG. It is determined that the timing at which the pattern can be generated by the first inkjet nozzle 1031 and the second inkjet nozzle 1041 has arrived, and the first inkjet 1032 and the second inkjet 1042 are controlled to generate the pattern. Further, the pattern imaging unit 1051 determines that it is time to image the drawn pattern when the limit switch 909 changes from the off state to the on state after the pattern is generated, that is, when the state of FIG. 30 changes to the state of FIG. 29. Then, the image is taken by the built-in camera.

Next, the pattern imaging unit 1051 will be described in detail.

FIG. 32 is a block diagram showing an example of the hardware configuration of the pattern imaging unit 1051. The pattern imaging unit 1051 of this example is composed of a camera 10511, an MPU 10512, a ROM 10513, a RAM 10514, an interface circuit 10515, a communication circuit 10516, and a bus 10517 that connects them to each other. Of these, the camera 10511, ROM 10513, RAM 10514, and communication circuit 10516 have the same functions as the camera 9021, ROM 9023, RAM 9024, and communication circuit 9026 described with reference to FIG. 25. The interface circuit 10515 is connected to a limit switch 909, 910, an illuminator 914, a first inkjet 1032, and a second inkjet 1042 through a signal line (not shown).

FIG. 33 is a flowchart showing an example of the processing executed by the MPU 10512. Hereinafter, the process executed by the MPU 10512 will be described with reference to FIGS. 29 to 33.

The MPU 10512 reads the state of the limit switch 909 through the interface circuit 10515 and monitors whether or not the limit switch 909 is turned off (step S1001). As described above, in the state shown in FIG. 29, the object identifier registration device 1000 has the limit switch 909 turned on, and when it is placed on the object to generate a pattern and the lid 905 is pushed downward, One end of the movable mirror 908 is pushed down by the tips of the first and second inkjet nozzles 1031 and 1041, and the limit switch 909 changes from the on state to the off state. When the MPU 10515 detects that the limit switch 909 has changed to the off state, it detects whether or not the limit switch 910 has changed to the on state (step S1002).

When the limit switch 910 is turned on by pushing the lid 905 further to a position where the user cannot push it down any further, the MPU 10515 determines that it is possible to generate a pattern on the object, and the pattern is changed. The generation is performed (step S1003). Specifically, the MPU 10512 first issues a command to the first inkjet 1032, and after a predetermined time elapses, issues a command to the second inkjet 1042. The first inkjet 1032 draws a large point on an object by ejecting a predetermined amount of ink supplied from the ink cartridge 1033 from the tip of the first inkjet nozzle 1031 in response to a command. Further, the second inkjet 1042 ejects a predetermined amount of ink supplied from the ink cartridge 1043 from the tip of the second inkjet nozzle 1041 in response to a command, so that the ink is small within the points drawn on the object. Draw a point. Then, the MPU 10512 monitors that the limit switch 909 changes to the ON state (step S1004). On the other hand, the MPU 10512 has a lower lid 905 when the limit switch 909 is returned to the on state after the limit switch 909 is turned off and before the limit switch 910 is turned on (YES in step S1005). Is determined to have risen on the way, and the process returns to step S1001.

Next, when the MPU 10512 detects that the limit switch 909 is turned on in step S1004, it issues a command to the camera 10511 to take an image (step S1006). The camera 10511 captures an image of the pattern generated on the object in response to this command. The state of the object identifier registration device 1000 at this time is as shown in FIG. 29, and the camera 10511 captures an image of the points drawn on the object as an image from directly above through the movable mirror 908 and captures the image. The obtained image data is stored in a predetermined area of the RAM 10514.

Next, the MPU 10512 analyzes the image data stored in the RAM 10514 and extracts the object identifier from the image of the drawn points (step S1007). Specifically, the MPU 10512 extracts an object identifier from the circular random pattern 1061 as shown in FIG. 31 using an iris recognition algorithm, similarly to the first embodiment shown in FIG. Next, the MPU 10512 transmits the extracted individual identifier to an external registration device (not shown) through the communication circuit 10516 (step S1008). Then, the MPU 10512 returns to the process of step S1001 in preparation for the generation and imaging of the next pattern.

As described above, according to the present embodiment, the pattern imaging unit 1051 detects that a pattern is generated on the object by the first and second inkjet nozzles 1031 and 1041 and performs imaging, so that the pattern is generated. It is possible to prevent a discrepancy from the imaging.

Further, according to the present embodiment, it is possible to generate an annular pattern in which minute particles are randomly distributed on an object. Further, when the asymmetric pattern as shown in FIG. 31 is used, the orientation of the pattern at the time of shooting can be determined, so that the alignment and orientation estimation at the time of collation become unnecessary, and the collation processing can be speeded up.

Further, according to the present embodiment, the pattern generated on the object can be captured as an image from directly above the object.

Further, according to the present embodiment, the object identifier can be extracted from the captured pattern and registered in an external storage device.

In the present embodiment, the movable mirror 908 is used, but the movable mirror 908 may be omitted. In this case, as in FIGS. 27 and 28 showing the modified examples of the second embodiment, for example, the inkjet nozzle 1031 and the inkjet nozzle 1041 have a fixing plate so that the tips thereof are located above the center of the lower surface of the container 904. Attach it diagonally to 913. Further, the pattern imaging unit 902 is fixed to the container 904 so that the imaging direction of the built-in camera faces the center of the lower surface of the container 904. According to such a configuration, the movable mirror 908 is omitted, so that the configuration is simplified.

Further, if there is no problem even if a part of the inkjet nozzle 1031 and the inkjet nozzle 1041 is captured in the image of the drawn points, a state in which the tips of the inkjet nozzle 1031 and the inkjet nozzle 1041 are located near the object (for example, FIG. The image may be taken in the state of 30).

In this embodiment, an inkjet is used, but instead of an inkjet, an ink dropping device capable of dropping an appropriate amount (for example, one drop) of ink, an ink ejector capable of spraying an appropriate amount of ink, or an ink is used. You may use two sets of applicators that can apply an appropriate amount by brushing. Alternatively, instead of the inkjet, a laser machine that forms an annular random pattern on the object may be used.

Further, in the present embodiment, each time a pattern is generated and the image is captured, the object identifier is extracted and registered in the storage device, but the captured image is temporarily stored in a memory (RAM or the like) and the object is later stored. The identifier may be extracted and registered in the storage device. Further, in the present embodiment, the object identifier is registered in the storage device, but instead of or in addition to the object identifier, the image of the captured pattern may be registered in the storage device.

[11th Embodiment]
In this embodiment, another embodiment in which the object identifier registration device 600 according to the sixth embodiment is more embodied will be described.

With reference to FIG. 34, in the object identifier registration device 1100 according to the eleventh embodiment of the present invention, the ink pen 901 of the object identifier registration device 900 according to the ninth embodiment described with reference to FIG. 22 is stamped 1101. The pattern imaging unit 902 is replaced with the pattern imaging unit 1102. In addition, a limit switch 1103 is newly provided. In addition, the same members as those of the object identifier registration device 900 according to the ninth embodiment are designated by the same reference numerals.

The stamp 1101 has a structure in which a felt material 1111 having a hollow columnar shape is fitted and fixed to one end side of a columnar core material 1113, and an ink tank 1112 is attached to an upper core material portion of the felt material 1111. By fixing the other end side of the core member 1113 to the lower surface of the fixing plate 913, the core member 1113 is configured to move up and down in conjunction with the movement of the lid 903. Further, the fiberscope 1114 extending from the pattern imaging unit 1102 penetrates from the side surface of the upper end to the lower end of the core material 1113, and the image acquired by the lens attached to the tip portion is transmitted to the pattern imaging unit 1102. There is.

The ink tank 1112 stores ink mixed with fine particles. As the fine particles, fine particles such as metal powder and glass powder, taggants described in Patent Documents 1 to 3 and the like can be used. It is desirable that the fine particles are particles having different reflection characteristics from the material (excluding the fine particles) constituting the ink. Further, it is desirable that the fine particles are non-uniformly contained in the ink. The ink filled in the ink tank 1132 always soaks into the felt material 1131 due to the capillary phenomenon.

When the lid 905 is not pressed from above, the lower surface of the felt material 1111 is at a position (standby position) away from the bottom surface of the container 904. Further, in this state, the upper surface of the fixing plate 913 is in contact with the limit switch 1103, and the limit switch 1103 is in the ON state. On the other hand, in the state shown in FIG. 35 in which the lid 905 is pressed from above, the lower surface of the felt material 1111 that moves up and down integrally with the lid 905 is at a position (generation position) that reaches the lower surface of the container 904. As a result, when the container 904 is placed on the object, a pattern is generated on the object by the ink exuding from the lower surface of the hollow cylindrical felt material 1111. As described above, since fine particles are non-uniformly mixed in the ink, the pattern to be drawn includes minute particles (represented by black circles in the figure) as schematically shown in FIG. It becomes an annular pattern contained at random positions. When the lid 905 is lowered until the lower surface of the felt material 1111 comes into contact with the surface of the object, the limit switch 910 changes from the off state to the on state.

Next, the operation of the object identifier registration device 1100 will be described with reference to FIGS. 34 and 35.

When the user uses the object identifier registration device 1100 to generate a pattern for individual identification on an object, the user places the housing 903 on the object for which the pattern is to be generated so that the lower surface of the container 904 is in contact with the object. At this time, the object identifier registration device 1100 is in a state as shown in FIG. 34, the limit switch 1103 is in the on state, and the limit switch 910 is in the off state.

When the user pushes down the lid 905 of the housing 903 placed on the object, the state gradually transitions from the state shown in FIG. 34 to the state shown in FIG. 35. Then, when the user pushes down the lid 905 to a position where it cannot be pushed down any further, the lower surface of the felt material 1111 of the stamp 1101 that descends in conjunction with the lid 905 is pressed against the object, and an annular pattern is formed on the object. Will be generated. Further, since the fixing plate 913 that descends in conjunction with the lid 905 comes into contact with the limit switch 910, the limit switch 910 changes to the ON state.

Next, when the user releases the force pushing down the lid 905, the repulsive force of the elastic body 911 causes the lid 905 to rise toward the original position. In the process, the limit switch 910 first changes from the on state to the off state. When the lid 905 is further raised to reach the state shown in FIG. 34, the limit switch 1103 changes from the off state to the on state.

The pattern imaging unit 1102 detects the generation of the pattern based on the state of the limit switches 910 and 1103, and also detects the timing of imaging the generated pattern. For example, when the limit switch 1103 changes from the on state to the off state and then the limit switch 910 changes from the off state to the on state, that is, when the pattern imaging unit 1102 changes from the state of FIG. 34 to the state of FIG. 35, It is determined that the pattern is generated by the stamp 1101. Further, the pattern imaging unit 1102 said that when the limit switch 1103 changed from the off state to the on state after the pattern was generated, that is, when the state of FIG. 35 changed to the state of FIG. 34, it was time to image the generated pattern. Judgment is made, and an image of the generated pattern is captured by the built-in camera through the fiberscope 1113.

Next, the pattern imaging unit 1102 will be described in detail.

FIG. 37 is a block diagram showing an example of the hardware configuration of the pattern imaging unit 1102. The pattern imaging unit 1102 of this example is composed of a camera 11021, an MPU 11022, a ROM 11023, a RAM 11024, an interface circuit 11025, a communication circuit 11026, and a bus 11027 that connects them to each other. Of these, ROM 11023, RAM 11024, interface circuit 11025, communication circuit 11026, and bus 11027 have the same functions as ROM 9023, RAM 9024, interface circuit 9025, communication circuit 9026, and bus 9027 described with reference to FIG. 25.

As the camera 11021, for example, a camera using a CCD image sensor or a CMOS image sensor can be used. A fiberscope 1114 is connected to the camera 11021, and the structure is such that the image acquired by the fiberscope 1114 is captured by the camera 11021.

FIG. 38 is a flowchart showing an example of the processing executed by the MPU 11022. Hereinafter, the process executed by the MPU 11022 will be described with reference to FIGS. 34 to 38.

The MPU 11022 reads the state of the limit switch 1103 through the interface circuit 11025 and monitors whether or not it is turned off (step S1101). As described above, in the state of the object identifier registration device 1100 shown in FIG. 34, the limit switch 1103 is in the on state, and when the object is placed on the object and the lid 905 is pushed downward to generate a pattern, Since the fixing plate 913 is lowered, the limit switch 1103 changes from the on state to the off state. When the MPU 11022 detects that the limit switch 1103 has changed to the off state, it detects whether or not the limit switch 910 has changed to the on state (step S1102).

When the limit switch 910 is turned on by pushing the lid 905 further to a position where the user cannot push it down any further, the MPU 11022 has the felt material 1111 of the stamp 1101 built in the container 904 on the surface of the object. It is determined that the pattern is drawn on the object by the ink that is pressed and flows out from the felt material 1111. Next, the limit switch 1103 is monitored to change to the ON state (step S1103). On the other hand, the MPU 11022 is drawn by the stamp 1101 when the limit switch 1103 is returned to the on state after the limit switch 1103 is turned off and before the limit switch 910 is turned on (YES in step S1104). It is determined that the lid 905 has been returned without any trouble, and the process returns to step S1101.

Next, when the MPU 11022 detects that the limit switch 1103 is turned on in step S1103, it issues a command to the camera 11021 to take an image (step S1105). The camera 11021 captures an image acquired by the fiberscope 1114 in response to this command. When the illuminance is insufficient, the camera 11021 temporarily turns on the illuminator 914 at the time of shooting through the interface circuit 11025. The state of the object identifier registration device 1100 at the time of shooting is as shown in FIG. 34, and the camera 11021 directly above the annular image as shown in FIG. 36 drawn on the object with the stamp 1101. The image data obtained by imaging is stored in a predetermined area of the RAM 11024.

Next, the MPU 11022 analyzes the image data stored in the RAM 11024 and extracts the object identifier from the drawn circular image (step S1106). Specifically, MPU11022 uses an iris recognition algorithm to extract an object identifier from a circular random pattern as shown in FIG. 36, similar to the first embodiment shown in FIG. Next, the MPU 11022 registers the data including the extracted individual identifier in an external storage device (not shown) through the communication circuit 11026 (step S1107). The MPU 11022 may include information stored in the ROM 11023 or the RAM 11024 in the transmission data, for example, identification information of the object identifier registration device 1100 and attribute information such as the current time. Then, the MPU 11022 returns to the process of step S1101 in preparation for the next imaging.

As described above, according to the present embodiment, since the pattern imaging unit 1102 detects that the pattern is generated on the object by the stamp 1101 and performs the imaging, there is a discrepancy between the pattern generation and the imaging. Can be prevented.

Further, according to the present embodiment, it is possible to generate an annular pattern on the object.

Further, according to the present embodiment, the pattern generated on the object can be imaged from directly above the object.

Further, according to the present embodiment, the object identifier can be extracted from the captured pattern and registered in an external storage device.

In the present embodiment, the annular pattern is generated, but by changing the shape of the lower surface of the felt material 1111 of the stamp 1101, a pattern having a shape other than the annular pattern can be generated.

Further, in the present embodiment, the felt material 1111 is used, but the felt material is not limited as long as it is a member capable of forming a pattern with ink by pressing it against an object, and a sponge, cloth or the like may be used.

Further, in the present embodiment, each time a pattern is generated and the image is captured, the object identifier is extracted and registered in the storage device, but the captured image is temporarily stored in a memory (RAM or the like) and the object is later stored. The identifier may be extracted and registered in the storage device. Further, in the present embodiment, the object identifier is registered in the storage device, but instead of or in addition to the object identifier, the image of the captured pattern may be registered in the storage device.

[Twelfth Embodiment]
Referring to FIG. 39, the object identifier extraction device 1200 according to the present embodiment has an imaging unit 1201 and an extraction unit 1202.

The imaging unit 1201 has a function of imaging an object having a random pattern processed on its surface. The extraction unit 1202 has a function of extracting features of an annular image region including a random pattern as identification information of the object from the images of the object captured by the image pickup unit 1201.

As shown in FIG. 18, for example, the object identifier extraction device 1200 includes an information processing device 740 having an arithmetic processing unit 741 such as one or more microprocessors, a storage unit 742 such as a memory or a hard disk, and a camera 743, and a program. It can be realized with 744. The information processing device 740 may be, for example, a smartphone. The program 744 is read into the memory from an external computer-readable recording medium at the time of starting up the information processing device 740, and controls the operation of the arithmetic processing unit 741 so that the imaging unit 1201 is placed on the arithmetic processing unit 741. , The extraction unit 1202 and the like are realized.

FIG. 40 is a flowchart showing the procedure of the object identifier extraction method executed by using the object identifier extraction device 1200. Hereinafter, the object identifier extraction method according to the present embodiment will be described with reference to FIGS. 39 and 40.

First, the imaging unit 1201 of the object identifier extraction device 1200 acquires an image of a random pattern processed on the object (step S1201). Next, the extraction unit 1202 extracts the feature of the annular image region including the random pattern from the image of the object acquired by the imaging unit 1201 as the identification information of the object (step S1202).

As described above, according to the present embodiment, the technology cultivated in iris recognition can be applied to the certification of objects such as industrial products and commodities.

The reason is that the extraction unit 1202 extracts the feature of the annular image region including the random pattern from the image of the object acquired by the imaging unit 1201 as the identification information of the object, and the annular image including the random pattern. This is because the image area is an image area similar to the iris.

Although the present invention has been described above with reference to some embodiments, the present invention is not limited to the above embodiments, and various additions and changes are possible. Various modifications that can be understood by those skilled in the art can be made to the structure and details of the present invention within the scope of the present invention.

It should be noted that the present invention enjoys the benefit of priority claim based on the patent application of Japanese Patent Application No. 2015-183624, which was filed for patent on September 17, 2015 in Japan, and is described in the patent application. All contents are included in this specification.

The present invention can be used in the field of identifying and collating the identities of objects such as industrial products and commodities.

Some or all of the above embodiments may also be described, but not limited to:
[Appendix 1]
An imaging unit that captures an object with a random pattern on the surface,
An extraction unit that extracts features of an annular image region including the random pattern from the image of the captured object as identification information of the object.
Object identifier extractor including.
[Appendix 2]
The random pattern is circular,
The object identifier extraction device according to Appendix 1.
[Appendix 3]
A part of the random pattern is processed into a point-like region that is not a random pattern.
The extraction unit performs preprocessing to cut out an image in a circle having a predetermined length as a radius from the center of the point-shaped region in the image of the random pattern as the annular image region.
The object identifier extraction device according to Appendix 1.
[Appendix 4]
The random pattern is substantially circular and
The extraction unit performs preprocessing to generate the annular image region by generating a circular image centered on the center of gravity of the random pattern image on the random pattern image.
The object identifier extraction device according to Appendix 1.
[Appendix 5]
The extraction unit generates an image of a point centered on the center of gravity of the image of the random pattern on the image of the random pattern, and is a circle having a predetermined length as a radius from the center of the point in the image of the random pattern. Preprocessing is performed to cut out the image inside as the annular image area.
The object identifier extraction device according to Appendix 1.
[Appendix 6]
An image of an object with a random pattern processed on the surface is imaged.
From the image of the captured object, the feature of the annular image region including the random pattern is extracted as the identification information of the object.
Object identifier extraction method.
[Appendix 7]
The random pattern is circular,
The object identifier extraction method according to Appendix 6.
[Appendix 8]
A part of the random pattern is processed into a point-like region that is not a random pattern.
In the extraction, preprocessing is performed to cut out an image in a circle having a predetermined length as a radius from the center of the point-shaped region in the image of the random pattern as the annular image region.
The object identifier extraction method according to Appendix 6.
[Appendix 9]
The random pattern is substantially circular and
In the extraction, a preprocessing for generating the annular image region is performed by generating a circular image centered on the center of gravity of the random pattern image on the random pattern image.
The object identifier extraction method according to Appendix 6.
[Appendix 10]
In the extraction, an image of a point centered on the center of gravity of the image of the random pattern is generated on the image of the random pattern, and in the image of the random pattern, within a circle having a predetermined length as a radius from the center of the point. Preprocessing is performed to cut out the image of
The object identifier extraction method according to Appendix 6.
[Appendix 11]
An imaging unit that captures an object with a random pattern on the surface,
An extraction unit that extracts features of an annular image region including the random pattern from the image of the captured object as identification information of the object.
A determination unit that compares the identification information of the object with the registered object identification information stored in the storage unit, and determines the identification / collation of the object based on the comparison result.
Identification collation device including.
[Appendix 12]
An image of an object with a random pattern processed on the surface is imaged.
From the image of the captured object, the feature of the annular image region including the random pattern is extracted as the identification information of the object.
The object identification information is compared with the registered object identification information stored in the storage unit, and the object identification / collation is determined based on the comparison result.
Identification collation method.
[Appendix 13]
Computer,
An imaging unit that captures an object with a random pattern on the surface,
An extraction unit that extracts features of an annular image region including the random pattern from the image of the captured object as identification information of the object.
A program to make it work.
[Appendix 14]
A processing part that processes a random pattern on the surface of an object,
An imaging unit that captures an object with the random pattern processed,
An extraction unit that extracts features of an annular image region including the random pattern from the image of the captured object as identification information of the object.
Object identifier registration device including.
[Appendix 15]
A registration unit that registers the object identification information in the storage unit as the registered object identification information.
Have
The object identifier registration device according to Appendix 14.
[Appendix 16]
The processing portion is configured to be movable from a standby position to a generation position where the random pattern is generated.
The imaging unit detects the generation of the pattern by detecting that the processing unit has moved from the standby position to the generation position.
The object identifier registration device according to Appendix 14 or 15.
[Appendix 17]
The imaging unit performs imaging at the timing when the processing unit moves from the generation position to the standby position.
The object identifier registration device according to Appendix 16.
[Appendix 18]
It has a movable mirror provided in the middle of the movement path of the processed portion and guides the processed image of the random pattern to the imaging unit.
The object identifier registration device according to any one of Supplementary note 14 to 17.
[Appendix 19]
It has a fiberscope that captures an image of the processed random pattern from the vicinity of the processed portion and guides it to the imaging portion.
The object identifier registration device according to any one of Appendix 14 to 17.
[Appendix 20]
The processing unit processes the random pattern by applying a liquid containing fine particles to the object.
The object identifier registration device according to any one of Supplementary note 14 to 19.
[Appendix 21]
The processed portion is a pen.
The object identifier registration device according to any one of Supplementary note 14 to 19.
[Appendix 22]
The processed portion is a means for spraying liquid ink particles.
The object identifier registration device according to any one of Supplementary note 14 to 19.
[Appendix 23]
The means for spraying the liquid ink particles is an inkjet.
The object identifier registration device according to Appendix 22.
[Appendix 24]
The inkjet is a first inkjet that generates a circular first pattern by spraying ink containing fine particles onto the object, and the first inkjet that sprays ink that does not contain fine particles. It has a second inkjet that produces a second pattern in a circular shape smaller than the pattern on the first pattern.
The object identifier registration device according to Appendix 23.
[Appendix 25]
The processed portion is a stamp.
The object identifier registration device according to any one of Supplementary note 14 to 19.
[Appendix 26]
The shape of the random pattern processed on the object by the stamp is annular.
The object identifier registration device according to Appendix 25.
[Appendix 27]
The extraction unit extracts a feature amount depending on the distribution of the grains as an object identifier of the object from the image of the random pattern obtained by imaging.
The object identifier registration device according to any one of Supplementary note 14 to 26.
[Appendix 28]
The processing unit is a laser processing machine that generates a random pattern on the object.
The object identifier registration device according to any one of Supplementary note 14 to 19.
[Appendix 29]
Random pattern is processed on the surface of the object
An image of an object processed with the random pattern is imaged.
From the image of the captured object, the feature of the annular image region including the random pattern is extracted as the identification information of the object.
Object identifier registration method.
[Appendix 30]
The object identification information is registered in the storage unit as the registered object identification information.
The object identifier registration method according to Appendix 29.
[Appendix 31]
It has an object identifier registration device and an identification collation device.
The object identifier registration device is
A processing part that processes a random pattern on the surface of an object,
An imaging unit that captures an object with the random pattern processed,
An extraction unit that extracts features of an annular image region including the random pattern from the image of the captured object as identification information of the object.
A registration unit that registers the object identification information in the storage unit as registered object identification information,
Including
The identification collation device is
An imaging unit that captures an object with a random pattern on the surface,
An extraction unit that extracts features of an annular image region including the random pattern from the image of the captured object as identification information of the object.
A determination unit that compares the identification information of the object with the registered object identification information stored in the storage unit, and determines the identification / collation of the object based on the comparison result.
including,
Object identifier management system.
[Appendix 32]
It has an object identifier registration process and an identification collation process.
In the object identifier registration step,
Random pattern is processed on the surface of the object
An image of an object processed with the random pattern is imaged.
From the image of the captured object, the feature of the annular image region including the random pattern is extracted as the identification information of the object.
The identification information of the object is registered in the storage unit as the identification information of the registered object, and is registered.
In the identification collation step,
An image of an object with a random pattern processed on the surface is imaged.
From the image of the captured object, the feature of the annular image region including the random pattern is extracted as the identification information of the object.
The object identification information is compared with the registered object identification information stored in the storage unit, and the object identification / collation is determined based on the comparison result.
Object identifier management method.
[Appendix 33]
An identification device that identifies using an object identifier extracted by the object identifier extraction device according to any one of Supplementary note 1 to 5.
[Appendix 34]
A collation device that performs collation using the object identifier extracted by the object identifier extraction device according to any one of Appendix 1 to 5.
[Appendix 35]
An identification method for identifying using an object identifier extracted by the object identifier extraction method according to any one of Supplementary note 6 to 10.
[Appendix 36]
A collation method for collating using the object identifier extracted by the object identifier extraction method according to any one of Appendix 6 to 10.
[Appendix 37]
The size of the circular, amorphous or substantially circular random pattern is similar to that of a human iris.
The device or method according to any one of Supplementary notes 1 to 36.
[Appendix 38]
The photographing unit has a conversion lens that converts an image of the circular, amorphous, or substantially circular random pattern into a size similar to that of a person's eye color existing at the same position as the random pattern.
The device or method according to any one of Supplementary notes 1 to 36.

100 ... Object identifier extraction device 101 ... Imaging unit 102 ... Extraction unit 110 ... Object 111 ... Random pattern 120 ... Object identifier 200 ... Object identifier extraction device 201 ... Imaging unit 202 ... Extraction unit 210 ... Object 211 ... Random pattern 212 ... Dot pattern Area 220 ... Object identifier 213 ... Center of point-shaped area 214 ... Circle 300 ... Object identifier extraction device 301 ... Imaging unit 302 ... Extraction unit 310 ... Object 311 ... Random pattern 320 ... Object identifier 312 ... Circular image 400 ... Object Identifier extraction device 401 ... Imaging unit 402 ... Extraction unit 410 ... Object 411 ... Random pattern 420 ... Object identifier 413 ... Circle center 413 ... Circle 500 ... Object identifier registration device 501 ... Imaging unit 502 ... Extraction unit 503 ... Processing unit 510 ... Object 511 ... Random pattern 520 ... Object identifier 600 ... Object identifier registration device 601 ... Imaging unit 602 ... Extraction unit 603 ... Processing unit 604 ... Registration unit 610 ... Object 611 ... Random pattern 620 ... Storage device 700 ... Identification collation device 701 ... Imaging Unit 702 ... Extraction unit 703 ... Judgment unit 710 ... Object 711 ... Random pattern 720 ... Storage device 740 ... Information processing device 741 ... Arithmetic processing unit 742 ... Storage unit 743 ... Camera 744 ... Program 800 ... Object identifier management system 801 ... Object identifier Registration device 802 ... Identification and judgment device 810 ... Object 811 ... Random pattern 820 ... Storage device 830 ... Judgment result 900 ... Object identifier registration device 901 ... Ink pen 902 ... Pattern image pickup unit 903 ... Housing 904 ... Container 905 ... Lid 906 ... Light removal Window 908 ... Movable mirror 908a ... Shaft 908b ... Tension spring 909 ... Limit switch 910 ... Limit switch 911 ... Elastic body 912 ... Connecting material 913 ... Fixed plate 914 ... Illuminator 915 ... Limit switch 1000 ... Individual identification device 1011 ... Conversion Lens 1031 ... First inkjet nozzle 1032 ... First inkjet 1033 ... First first ink cartridge 1041 ... Second inkjet nozzle 1042 ... Second inkjet 1043 ... Second first ink cartridge 1052 ... Pattern Imaging unit 1100 ... Individual identification device 1102 ... Pattern imaging unit 1103 ... Limit switch 1111 ... Felt material 1112 ... Ink tank 1113 ... Core material 1114 ... Fiber scope 1200 ... Object identifier extraction device 1201 ... Imaging unit 1202 ... Extraction unit

Claims (4)

A processing part that processes a random pattern on the surface of an object,
An imaging unit that captures an object with the random pattern processed,
An extraction unit that extracts features of an annular image region including the random pattern from the image of the captured object as identification information of the object.
Only including,
The processed portion is a means for spraying liquid ink particles.
The means for spraying the liquid ink particles is an inkjet.
The inkjet includes a first inkjet that generates a circular first pattern by spraying ink containing fine particles onto the object, and the first inkjet that sprays ink that does not contain fine particles. It has a second inkjet that produces a second pattern in a circular shape smaller than the pattern on the first pattern.
Object identifier registration device. Random pattern is processed on the surface of the object
An image of an object processed with the random pattern is imaged.
From the image of the captured object, the feature of the annular image region including the random pattern is extracted as the identification information of the object .
In the processing, the first inkjet is used to spray ink containing fine particles onto the object to generate a circular first pattern, and the second inkjet is used to spray the fine grains. A second pattern having a circular shape smaller than the first pattern is generated on the first pattern by spraying ink that does not contain the above.
Object identifier registration method. It has an object identifier registration device and an identification collation device.
The object identifier registration device is
A processing part that processes a random pattern on the surface of an object,
An imaging unit that captures an object with the random pattern processed,
An extraction unit that extracts features of an annular image region including the random pattern from the image of the captured object as identification information of the object.
A registration unit that registers the object identification information in the storage unit as registered object identification information,
Including
The identification collation device is
An imaging unit that captures an object with a random pattern on the surface,
An extraction unit that extracts features of an annular image region including the random pattern from the image of the captured object as identification information of the object.
A determination unit that compares the object identification information with the registered object identification information stored in the storage unit, and determines the object identification / collation based on the comparison result.
Only including,
The processed portion is a means for spraying liquid ink particles.
The means for spraying the liquid ink particles is an inkjet.
The inkjet includes a first inkjet that generates a circular first pattern by spraying ink containing fine particles onto the object, and the first inkjet that sprays ink that does not contain fine particles. It has a second inkjet that produces a second pattern in a circular shape smaller than the pattern on the first pattern.
Object identifier management system. It has an object identifier registration process and an identification collation process.
In the object identifier registration step,
Random pattern is processed on the surface of the object
An image of an object processed with the random pattern is imaged.
From the image of the captured object, the feature of the annular image region including the random pattern is extracted as the identification information of the object.
The object identification information is registered in the storage unit as the registered object identification information.
In the identification collation step,
An image of an object with a random pattern processed on the surface is imaged.
From the image of the captured object, the feature of the annular image region including the random pattern is extracted as the identification information of the object.
The identification information of the object is compared with the registration information of the registered object stored in the storage unit, and the identification / collation of the object is determined based on the comparison result .
In the processing, the first inkjet is used to spray ink containing fine particles onto the object to generate a circular first pattern, and the second inkjet is used to spray the fine grains. An object identifier management method for generating a second pattern having a circular shape smaller than the first pattern on the first pattern by spraying ink that does not contain the above .

Images