WO2021157984A1

WO2021157984A1 - Container recognition module and method for packaging sealer

Info

Publication number: WO2021157984A1
Application number: PCT/KR2021/001314
Authority: WO
Inventors: 김영민; 송재학; 박상현
Original assignee: (주) 세창실업; 주식회사 이에스피
Priority date: 2020-02-03
Filing date: 2021-02-01
Publication date: 2021-08-12
Also published as: KR102207019B1

Abstract

A container recognition module of a packaging sealer according to an embodiment recognizes an image of a container inserted into packaging machinery or a sealer, compares the image of the inserted container with an image of a genuine container, and, when the inserted container is approved as a genuine container, performs a packaging process. The container recognition module of a packaging sealer according to an embodiment, when the inserted container is not a genuine container, restricts use of the packaging machinery or the sealer to reduce the defect rate of a finished product, and improves the convenience of use of the module by identifying whether or not the recognized container is genuine, regardless of the insertion direction of the container to be recognized.

Description

Container recognition module and method of sealing machine for packaging

It relates to a container recognition module and method, and more specifically, to a container recognition module and method for a packaging sealer that is mounted on a packaging sealer and compares images of a genuine container and a container put into the sealer to confirm whether the inserted container is genuine.

Unless otherwise indicated herein, the material described in this section is not prior art to the claims of this application, and inclusion in this section is not an admission that it is prior art.

A packaging machinery is a machine for packaging and sealing articles, and refers to an automatic packaging machinery. When referring to individual machines, they are called packaging machines and automatic packaging machines. Packaging machines are generally classified into full-automatic packaging machines and semi-automatic packaging machines according to the supply method of packaging target products and packaging materials. In addition, according to the type of packaging, it is classified into remodeling or interior machines and exterior machines, respectively. Sealing machines are used in various places from production plants to homes.

Conventional packaging machines perform packaging or sealing immediately after recognition when the input of a product to be sealed is recognized.

※ Prior art literature

Patent Literature 1. Korean Patent Publication No. 10-2018-0050635 (2018.05.15)

Provided is a container recognition module and method of a sealing machine for packaging that recognizes the image of a container put into a packaging machine or sealer, compares the image of the container with the genuine image, and then proceeds with the packaging process when the inserted container is approved as a genuine container.

In the embodiment, in order to limit the use of non-genuine containers by utilizing a dedicated container recognition module, image recognition is performed by checking whether the packaging container is genuine to check whether the packaging container is genuine when sealing the cover in the container for delivery and the automatic note packaging machine. Performs the authenticity recognition of the sealing machine container. For example, the packaging container is engraved with characters, patterns, and marks for recognizing genuine products, and after recognizing the shape engraved on the container by using a camera, the use of an automatic packaging machine is restricted if it is not a genuine product.

The recognition module according to the embodiment learns the shape of a mark, logo, trademark, etc. to be recognized from the camera or image recognition module, and based on the learned information, it is possible to determine whether a new container is genuine when a new container is inserted.

The container recognition module of the sealing machine for packaging according to an embodiment includes: a database for storing images of genuine containers and container image learning data; Recognition unit for recognizing the input of the container into the sealing machine for packaging; When the container recognition module is not in the container image learning mode, a learning unit for inputting the recognized container image to perform machine learning and learning the genuine container image and the recognized container image pattern; When the container recognition module is in the image comparison mode, a comparison unit for comparing the input container image and the genuine container image according to the learning result of the learning unit; and a control unit for performing a container packaging process according to the comparison result of the input container image and the genuine container image; includes

The container recognition method of the sealing machine for packaging according to another embodiment comprises the steps of: (A) storing the image of the genuine container and the container image learning data in the container recognition module of the sealing machine for packaging; (B) recognizing the input of the container into the sealing machine for packaging in the container recognition module of the sealing machine for packaging; (C) checking the operation mode in the container recognition module of the sealing machine for packaging; (D) when the container recognition module of the sealing machine for packaging is in the learning mode, inputting the recognized container image to machine learning, and learning the genuine container image and the recognized container image pattern; (E) when the container recognition module of the sealing machine for packaging is in the image comparison mode, comparing the input container image and the genuine container image according to the pattern learning result; and (F) performing a container packaging process according to the result of comparing the input container image with the genuine container image in the container recognition module of the sealing machine for packaging; includes

The container recognition module of the sealing machine for packaging according to the embodiment restricts the use of non-genuine containers to lower the defective rate of the finished product, and enables to check whether the recognition container is genuine regardless of the input direction of the container to be recognized, Improve the usability of the module.

It should be understood that the effects of the present invention are not limited to the above-described effects, and include all effects that can be inferred from the configuration of the invention described in the detailed description or claims of the present invention.

1 is a view showing a data processing configuration of a container recognition module of a sealing machine for packaging according to an embodiment;

2 is a view for explaining a feature extraction process of the container recognition module of the sealing machine for packaging according to the embodiment;

Figure 3 is a view showing a data processing process of the container recognition method of the sealing machine for packaging according to the embodiment;

4 is a view showing sectors for each region for comparing a recognized image and a genuine image for each region according to the embodiment;

Advantages and features of the present invention, and a method for achieving them will become apparent with reference to the embodiments described below in detail in conjunction with the accompanying drawings. However, the present invention is not limited to the embodiments disclosed below, but may be implemented in various different forms, and only these embodiments allow the disclosure of the present invention to be complete, and common knowledge in the art to which the present invention pertains. It is provided to fully inform the possessor of the scope of the invention, and the present invention is only defined by the scope of the claims. Like reference numerals refer to like elements throughout.

In describing the embodiments of the present invention, if it is determined that a detailed description of a well-known function or configuration may unnecessarily obscure the gist of the present invention, the detailed description thereof will be omitted. In addition, the terms to be described later are terms defined in consideration of functions in an embodiment of the present invention, which may vary according to intentions or customs of users and operators. Therefore, the definition should be made based on the content throughout this specification.

1 is a view showing a data processing configuration of a container recognition module of a sealing machine for packaging according to an embodiment. As used herein, the term 'module' should be construed to include software, hardware, or a combination thereof, depending on the context in which the term is used. For example, the software may be machine language, firmware, embedded code, and application software. As another example, the hardware may be a circuit, a processor, a computer, an integrated circuit, an integrated circuit core, a sensor, a Micro-Electro-Mechanical System (MEMS), a passive device, or a combination thereof.

Referring to FIG. 1 , the container recognition module of the sealing machine for packaging according to the embodiment includes a camera 10 , a database 15 , a recognition unit 20 , a learning unit 30 , a comparison unit 40 , and a control unit 50 . And it may be configured to include an output unit (60).

The database 15 stores an image of a genuine container and container image learning data. For example, the database 15 learns and stores the logo, genuine mark, trademark image, aspect ratio of the genuine container, etc. that appear on the genuine container, and the learning data of the recognized container image and the characteristic data of the container determined to be non-genuine. It stores a series of data necessary for image judgment and machine learning, etc.

The recognition unit 20 is a sensor for recognizing the input of a container into a sealing machine or a packaging machine, and is mounted on the container inlet side of the sealing machine or packaging machine to recognize that a container to be recognized is input, and creates an image with a camera after recognition is complete Send a request signal.

The camera 10 is mounted on the container inlet of the sealing machine according to the image generation request signal transmitted from the recognition unit 20 to generate an image of the container entering the sealing machine or the packaging machine, and the generated image is used by the learning unit 30 ) to pass

The learning unit 30 checks the operation mode of the container recognition module, and when the container recognition module is in the image learning mode, inputs the recognized container image to perform machine learning, and learns the genuine container image and the recognized container image pattern. In addition, the learning unit 30 generates learning data in which genuine image information and recognized image information are combined to model the characteristics of the image data recognized as learning data. For example, when generating learning data by overlapping the outermost edges of the genuine image and the recognized image, the learning unit 30 checks whether the size and aspect ratio of the two images match, and whether the objects inside the image overlap a certain level or more. figure out Also, learn the proportions and sizes of logos, genuine certification marks, and trademarks according to the genuine image and the recognized container image size.

When the container image comparison mode of the container recognition module starts, the comparison unit 40 compares the input container image with the genuine container image. It receives the characteristic information of the recognized image data from the learning unit, generates a primary comparison mask that extracts edges and characteristic points from the recognized image, and extracts the matching rate between the edge and the characteristic point of the genuine image and the recognized image.

Thereafter, the comparison unit 40 compares the recognized image and the genuine image by area when the match rate between the edge and the feature point of the genuine image and the recognized image is equal to or greater than the set value, and compares the minimum, maximum, average and standard values of the sector image match rate for each area. Calculate the deviation value. The comparator 40 determines that the difference between the minimum value and the maximum value of the calculated sector image matching rate is less than or equal to a preset value, and the standard deviation value is less than the preset value. If the difference between the average value and the maximum value and the difference between the average value and the minimum value is less than a certain range, the recognized image is approved as a genuine image.

The control unit 50 proceeds the container packaging process according to the comparison result of the input container image and the genuine container image.

The output unit 60 outputs the recognition state information and the comparison result of the put container and provides it to the user.

2 is a view for explaining a feature extraction process of the container recognition module of the sealing machine for packaging according to the embodiment.

The container recognition module of the packaging sealing machine according to the embodiment checks whether the packaging container is genuine when the cover is sealed in the delivery container and the automatic packaging machine in order to limit the use of the sealing machine or the packaging machine when it is not a genuine container. As shown in Figure 2, the comparison unit 40 of the container recognition module of the sealing machine for packaging according to the embodiment extracts the recognized edge (E1) of the surface of the container, whether the aspect ratio of the edge is the same as that of the genuine container Check it. Then, by extracting the characteristic areas (F1, F2, F3) in which the logo, genuine mark, brand mark, etc. are displayed on the surface of the container, it is determined whether the logo and mark included in the characteristic area match the genuine product.

Hereinafter, the container recognition method of the sealing machine for packaging will be described in turn. Since the action (function) of the container recognition method of the sealing machine for packaging according to the embodiment is essentially the same as the function of the container recognition module of the sealing machine for packaging, the description overlapping with FIGS. 1 and 2 will be omitted.

3 is a view showing a data processing process of the container recognition method of the sealing machine for packaging according to the embodiment.

The image of the genuine container and the container image learning data are stored in the container recognition module of the sealing machine for packaging, and in step S10, the recognition unit recognizes the input of the container into the sealing machine for packaging.

In step S20, the learning unit checks the operation mode in the container recognition module.

When the container recognition module of the sealing machine for packaging is in the image learning mode, in step S30, the recognized container image is input and machine learning is performed, and the genuine container image and the recognized container image pattern are learned. In the embodiment, step S30 may generate learning data in which genuine image information and recognized image information are combined to model the characteristics of the image data recognized as the learning data.

In step S40, machine learning is performed by inputting the recognized container image, and in step S50, it is determined whether the recognized container image learning is finished. When finished, the learning information is stored in step S60, and if the container image learning is not finished, it re-enters step S40.

If the container recognition module is in the image comparison mode in step S20, it goes to step S70 and puts the container into a sealing machine or a packaging machine. In step S80, a genuine image to be compared with the input container image is loaded, and in step S90, it is determined whether the container image and the genuine image match.

In the embodiment, in step S90, the characteristic information of the recognized image data is received from the learning unit, a primary comparison mask is generated that extracts the edges and the characteristic points from the recognized image, and the edge and the characteristic point of the genuine image and the recognized image are generated. Extract the match rate. In addition, when the coincidence rate between the edge and feature points of the genuine image and the recognized image is equal to or greater than the set value, the recognized image and the genuine image are compared by region to calculate the minimum, maximum, average, and standard deviation values of the sector image coincidence rate for each region. .

In the embodiment, when the difference between the minimum value and the maximum value of the calculated sector image coincidence rate is less than a preset value, the standard deviation value is less than the preset value, and the difference between the average value and the maximum value and the difference between the average value and the minimum value is less than a certain range, the recognized Approve the image as a genuine image. Thereafter, the container matching message is output in step S100 and the packaging process of the container approved as a genuine image is performed in step S110. If it is determined in step S90 that the container images do not match, a container mismatch message is output in step S95.

4 is a view showing sectors for each region for comparing a recognized image and a genuine image for each region according to the embodiment.

Referring to FIG. 4 , the container recognition module of the sealing machine for packaging according to the embodiment generates sectors S1 and S2 in which the recognized container image and the genuine image are divided by area, and compares each sector in the two images. The minimum value, maximum value, average value, and standard deviation value of the sector image matching rate for each area are calculated. In the embodiment, when the difference between the minimum value and the maximum value of the calculated sector image coincidence rate is less than a preset value, the standard deviation value is less than the preset value, and the difference between the average value and the maximum value and the difference between the average value and the minimum value is less than a certain range, the recognized Approve the container as genuine. In the embodiment, by calculating the coincidence rate of each sector for each region, only when the coincidence rate of the entire recognized container image is above a certain level, the genuine image is approved to prevent theft of logos and trademarks. Even the smallest difference can be detected.

The disclosed content is merely an example, and can be variously changed and implemented by those of ordinary skill in the art without departing from the gist of the claims claimed in the claims, so the protection scope of the disclosed content is limited to the specific It is not limited to an Example.

Claims

In the container recognition module of the sealing machine for packaging,

Database for storing images of genuine containers and container image learning data;

Recognition unit for recognizing the input of the container into the sealing machine for packaging;

When the container recognition module is not in the container image learning mode, a learning unit for inputting the recognized container image to perform machine learning and learning the genuine container image and the recognized container image pattern;

a comparison unit for comparing the input container image and the genuine container image according to the learning result of the learning unit when the container recognition module is in the image comparison mode; and

a control unit for performing a container packaging process according to the comparison result of the input container image and the genuine container image; Container recognition module of the sealing machine for packaging comprising a.
According to claim 1, wherein the learning unit; Is

Container recognition module for packaging sealing machine, characterized in that it generates learning data in which genuine image information and recognized image information are combined to model the characteristics of the image data recognized as the learning data.
The method of claim 1, wherein the comparator; Is

Receives feature information of the recognized image data from the learning unit, generates a primary comparison mask that extracts edges and feature points from the recognized image, Container recognition module of the sealing machine for packaging, characterized in that for calculating the coincidence rate.
The method of claim 3, wherein the comparator; Is

When the coincidence rate between the edge and feature points of the genuine image and the recognized image is equal to or greater than the set value, the recognized image and the genuine image are compared for each region to calculate the minimum, maximum, average, and standard deviation values of the sector image coincidence rate for each region Container recognition module of the sealing machine for packaging, characterized in that.
The method of claim 4, wherein the comparator;

A difference between the minimum value and the maximum value of the calculated sector image coincidence rate is less than or equal to a preset value, and a standard deviation value is less than a preset value. Container recognition module for packaging sealing machine, characterized in that when the difference between the average value and the maximum value and the difference between the average value and the minimum value is less than a certain range, the recognized image is approved as a genuine image.
In the container recognition method of the sealing machine for packaging,

(A) storing the image of the genuine container and the container image learning data in the container recognition module of the sealing machine for packaging;

(B) recognizing the input of the container into the sealing machine for packaging in the container recognition module of the sealing machine for packaging;

(C) checking the operation mode in the container recognition module of the sealing machine for packaging;

(D) when the container recognition module of the sealing machine for packaging is in the learning mode, inputting the recognized container image to machine learning, and learning the genuine container image and the recognized container image pattern;

(E) when the container recognition module of the sealing machine for packaging is in the image comparison mode, comparing the input container image and the genuine container image according to the pattern learning result; and

(F) performing a container packaging process according to the comparison result of the input container image and the genuine container image in the container recognition module of the sealing machine for packaging; A container recognition method of a sealing machine for packaging comprising a.
The method of claim 6, wherein the step (D); Is

A container recognition method for a packaging sealing machine, characterized in that it generates learning data in which genuine image information and recognized image information are combined to model the characteristics of the image data recognized as the learning data.
The method of claim 6, wherein the step of (E);

generating a primary comparison mask for receiving characteristic information of the recognized image data from the learning unit and extracting edges and characteristic points from the recognized image; and

extracting a coincidence rate between an edge and a feature point of the genuine image and the recognized image; A container recognition method of a sealing machine for packaging, characterized in that it comprises a.
The method of claim 8, wherein the step (E); Is

When the coincidence rate between the edge and feature points of the genuine image and the recognized image is equal to or greater than the set value, the recognized image and the genuine image are compared for each region to calculate the minimum, maximum, average, and standard deviation values of the sector image coincidence rate for each region A container recognition method of a sealing machine for packaging, characterized in that.
10. The method of claim 9, wherein the step (E); Is

A difference between the minimum value and the maximum value of the calculated sector image coincidence rate is less than or equal to a preset value, and a standard deviation value is less than a preset value. When the difference between the average value and the maximum value and the difference between the average value and the minimum value are less than a certain range, the recognized image is approved as a genuine image.