KR102540291B1 - Method and device for blockchain-based contest work management - Google Patents

Abstract

Disclosed are a method and a device for blockchain-based open contest entry-work management. The method for blockchain-based open contest entry-work management run by at least one server comprises the steps of: receiving, from a first user, information on open contest entry or award; receiving, from a second user, authentication about the information on open contest entry or award; determining whether an open contest participating work or an awarded work is made by plagiarism; and recording the information on open contest entry or award to a blockchain. At this time, the information on open contest entry or award is recorded to the blockchain only when the open contest participating work or the awarded work is determined not to be made by plagiarism, and is authentically submitted to the open contest hosted by or delegated to the second user or authenticated to be awarded from the open contest hosted by or delegated to the second user.

Description

Blockchain-based contest entry management method and device {METHOD AND DEVICE FOR BLOCKCHAIN-BASED CONTEST WORK MANAGEMENT}

The following embodiments relate to a blockchain-based method and device for managing entries for a contest that are used for contest introduction and contest career management platforms.

According to the data announced annually by the contest agency, the number of contests held annually as of 2021 is 13,000 to 14,000, with an average of more than 1,000 contests held per month. An average of 32 winners are created per contest, and when converted into the number of contests held, it can be seen that about 450,000 awards are generated annually in the contest field alone. Including areas where certification services can be applied (contests, hackathons, internships, supporters and volunteer activities, etc.) by expanding the scope to activities of a similar nature, a huge amount of awards/participation records are being created every year.

However, since the winner's history information is mainly in physical form, such as a certificate or plaque, there is no way for individuals to reconfirm it by themselves in case of loss, damage, or theft. In addition, in the reality that organizations hosting/supervising or acting on behalf of the contest do not have a separate management system, the person in charge of planning the contest is managing individual files, etc. there is. Furthermore, even if the contest listing is forged, the reality is that there is no way for a third party to easily verify the authenticity.

It is necessary to prepare a system and service that can prevent various unnecessary social costs such as damage to trust and occurrence of verification costs due to inefficient business structures and malicious deviant behaviors, and to reliably verify award history.

On the other hand, if a contest entry or award-winning work is a plagiarism work, it is necessary to prevent the work from being published online. Through this, plagiarism can be eradicated, and reliability can be guaranteed even when contest entries or award-winning works are posted online. In addition, there are cases in which a person or organization that supervises or acts on behalf of a contest wants to check whether or not plagiarism is present in the contest entries or prize-winning works. For example, the organization in charge of the contest may want to confirm whether the work is not plagiarized before awarding the contest entry subject to evaluation. In addition, the contest agency may want to increase the reliability of work by checking whether there are plagiarism in contest entries or prize-winning works managed by them and informing the contest host organization. Therefore, it is necessary to prepare a system capable of determining whether a contest entry or prize-winning work is plagiarized.

As background art related to the embodiments, Korean Registered Patent Publication No. KR 10-1983530 B1 discloses a registration system for storing copyrighted works in a block chain. The work providing server of the prior literature receives and stores the work, transmits the work requested from the user device to the user device as an effective transaction by cryptocurrency is performed, and transfers the proceeds obtained from the valid transaction to the right distribution information for each copyright. Based on this, it is distributed as cryptocurrency to the copyright holder. The blockchain network requests payment in cryptocurrency to receive the desired work from the work providing server, and as the payment is made, the requested work is provided from the work providing server and generated based on the transaction for the user device and work viewed. Verify the hash value. It is possible to support the simple transaction of subdivided copyright items such as secondary copyrights between copyright holders and consumers.

In addition, Republic of Korea Registered Patent Publication No. KR 10-2412511 B1 discloses a content management device and method using a blockchain DID. The method of the prior literature determines whether the content contains DID information, and if the content includes DID information, determines the authenticity of the content by managing a block containing the DID information through a blockchain network, and determines whether the content contains DID information. If the information is not included, search the reference content that most accurately matches the content through the Internet, set the second DID information, which is the DID information of the reference content, to the first DID information, which is the DID information of the content, and set the first DID information to the blockchain. register on the network Authenticity of the content can be determined on the blockchain network by restoring the unique feature using the content identification information and comparing the content and the unique feature to determine whether the content includes the unique feature.

However, in the prior literature, the contest participation or award history is recorded on a blockchain basis only when the contest entry or award-winning work is judged not to be plagiarism, and only when it has been submitted to or awarded in a contest hosted or run by a host organization or agency. It does not disclose a method and apparatus for doing so. In addition, the preceding documents do not disclose a method and apparatus for providing plagiarism based on artificial intelligence for contest entries or award-winning works to be verified at the request of a person or organization who supervises or acts on behalf of the contest. Furthermore, prior art documents do not disclose a method and apparatus for preparing training data for determining plagiarism of contest entries or prize-winning works based on artificial intelligence. Accordingly, the implementation of a technology for solving a technical problem that is not disclosed, suggested, or implied in the preceding documents is requested.

In the examples, only when the contest entry or award-winning work is judged not to be plagiarism, and it is verified that it has been submitted to or awarded in a contest hosted or run by a host organization or an agency, it is possible to record the contest participation or award career on a blockchain-based basis. It is intended to provide a method and apparatus.

Embodiments are intended to provide a method and apparatus for providing, based on artificial intelligence, plagiarism of contest entries or prize-winning works to be verified at the request of a person or organization that supervises or acts on behalf of the contest.

Embodiments are intended to provide a method and apparatus for preparing training data for determining plagiarism of contest entries or prize-winning works based on artificial intelligence.

In addition, the embodiments are intended to provide a method and apparatus for solving the problems of the technical field revealed in this specification.

A method for managing entries for a blockchain-based contest, performed by at least one server according to an embodiment, includes receiving contest entry or award information from a first user; and receiving authentication of the contest entry or award information from a second user, wherein the contest entry or award information includes a contest entry or award-winning work, and the authentication of the contest entry or award information is performed as described above. Including verification of whether the contest entries or award-winning works have really been submitted to the contest hosted or acted by the second user or won in the contest hosted or acted by the second user, the blockchain-based contest entry management method Determining whether the contest entries or award-winning works are plagiarized; And further comprising the step of recording the contest entry or award information in the blockchain, and it is determined that the contest participation or award-winning work is not plagiarism, and it is really entered into the contest organized or run by the second user, or the second user 2 Only when it is certified as an award winner in a contest organized or run by a user, the contest entry or award information can be recorded in the blockchain.

According to an embodiment, the blockchain-based method for managing contest entries is to apply each contest entry or award-winning work stored in a database linked to the server to pre-learned artificial intelligence, and each contest entry entry stored in the database. Alternatively, the method may further include obtaining embedding vectors of each of the award-winning works.

According to an embodiment, the step of determining whether the contest entry or award-winning work is plagiarized is the step of obtaining an embedding vector of the contest entry or award-winning work by applying the contest entry or award-winning work to the artificial intelligence. ; Determining a first similarity between each contest entry or award-winning work stored in the database and first plagiarism; transmitting each first similarity exceeding a first threshold and a first plagiarism status to the first user; and storing the contest entries or award-winning works in the database regardless of the first plagiarism result.

According to an embodiment, the blockchain-based method for managing entries for a contest may include receiving a contest entry or award-winning work subject to verification from the second user; obtaining an embedding vector of the contest entry or award-winning work to be verified by applying the contest entry or award-winning work to the artificial intelligence; determining a second degree of similarity between each contest entry or prize-winning work stored in the database and a second plagiarism; and transmitting the second similarity exceeding a second threshold and the second plagiarism status to the second user.

According to an embodiment, a block chain-based contest entry management apparatus includes at least one processor configured to receive contest entry or award information from a first user and authentication of the contest entry or award information from a second user. Including, the contest entry or award information includes a contest entry or award-winning work, and the authentication of the contest entry or award information is that the contest entry or award-winning work is indeed a contest organized or run by the second user. Including verification of whether it has been entered into or whether it has been awarded in a contest organized or run by the second user, the processor determines whether the contest entry or award-winning work is plagiarized, and puts the contest entry or award information on the blockchain Only when it is determined that the contest entry or award-winning work is not plagiarism, and it has been truly submitted to the contest organized or run by the second user or won in the contest hosted or run by the second user, The contest entry or award information can be recorded in the blockchain.

In the examples, only when the contest entry or award-winning work is judged not to be plagiarism, and it is verified that it has been submitted to or awarded in a contest hosted or run by a host organization or an agency, it is possible to record the contest participation or award career on a blockchain-based basis. A method and apparatus can be provided.

Embodiments may provide a method and apparatus for providing plagiarism based on artificial intelligence for a contest entry or award-winning work subject to verification at the request of a person or organization supervising or acting on behalf of the contest.

Embodiments may provide a method and apparatus for preparing training data for determining plagiarism of contest entries or prize-winning works based on artificial intelligence.

On the other hand, effects according to the embodiments are not limited to those mentioned above, and other effects not mentioned will be clearly understood by those skilled in the art from the description below.

1 is an exemplary diagram of a configuration of a system according to an embodiment.
2 is a diagram for explaining the operation of a system according to an embodiment.
3 is a flowchart illustrating a plagiarism determination operation according to an embodiment.
4 is a flowchart illustrating an operation of determining plagiarism according to a request from a contest organizer or agency according to an embodiment.
5 is a diagram for explaining a data augmentation operation for learning an artificial neural network according to an embodiment.
6 is a diagram for explaining learning of an artificial neural network according to an embodiment.
7 is an exemplary diagram of a configuration of a device according to an embodiment.

Hereinafter, embodiments will be described in detail with reference to the accompanying drawings. However, since various changes can be made to the embodiments, the scope of the patent application is not limited or limited by these embodiments. It should be understood that all changes, equivalents or substitutes to the embodiments are included within the scope of rights.

Specific structural or functional descriptions of the embodiments are disclosed for illustrative purposes only, and may be modified and implemented in various forms. Therefore, the embodiments are not limited to the specific disclosed form, and the scope of the present specification includes changes, equivalents, or substitutes included in the technical spirit.

Although terms such as first or second may be used to describe various components, such terms should only be construed for the purpose of distinguishing one component from another. For example, a first element may be termed a second element, and similarly, a second element may be termed a first element.

It should be understood that when an element is referred to as being “connected” to another element, it may be directly connected or connected to the other element, but other elements may exist in the middle.

Terms used in the examples are used only for the purpose of explanation and should not be construed as limiting. Singular expressions include plural expressions unless the context clearly dictates otherwise. In this specification, terms such as "include" or "have" are intended to designate that there is a feature, number, step, operation, component, part, or combination thereof described in the specification, but one or more other features It should be understood that the presence or addition of numbers, steps, operations, components, parts, or combinations thereof is not precluded.

The spatially relative terms "below", "beneath", "lower", "above", "upper", etc. It can be used to easily describe a component's correlation with other components. Spatially relative terms should be understood as encompassing different orientations of components in use or operation in addition to the orientations shown in the drawings. For example, if you flip a component that is shown in a drawing, a component described as "below" or "beneath" another component will be placed "above" the other component. can Thus, the exemplary term “below” may include directions of both below and above. Components may also be oriented in other orientations, and thus spatially relative terms may be interpreted according to orientation.

Unless defined otherwise, all terms used herein, including technical or scientific terms, have the same meaning as commonly understood by a person of ordinary skill in the art to which the embodiment belongs. Terms such as those defined in commonly used dictionaries should be interpreted as having a meaning consistent with the meaning in the context of the related art, and unless explicitly defined in the present application, they should not be interpreted in an ideal or excessively formal meaning. don't

In addition, in the description with reference to the accompanying drawings, the same reference numerals are given to the same components regardless of reference numerals, and overlapping descriptions thereof will be omitted. In describing the embodiment, if it is determined that a detailed description of a related known technology may unnecessarily obscure the gist of the embodiment, the detailed description will be omitted.

1 is an exemplary diagram of a configuration of a system according to an embodiment.

A system according to an embodiment includes at least one server 100; a first terminal 111; second terminal 112; a third terminal 113; A database 130 may be included. At least one server 100 and terminals 111 to 113 may be connected to a network and exchange data through wired or wireless communication. At least one server 100 and database 130 may be connected to a network and exchange data through wired or wireless communication.

At least one server 100 may be a server owned by a person or organization that provides services using the server; may be a cloud server; It may also be a peer-to-peer (p2p) set of distributed nodes. At least one server 100 may not be a single physical device, and may be composed of a plurality of physical devices. Hereinafter, at least one server 100 is abbreviated as server 100 for convenience of description.

The server 100 includes an arithmetic function that a normal computer has; save/reference function; input/output function; And it may be configured to perform all or part of the control function. The server 100 may include at least one artificial intelligence performing an inference function. The server 100 may be configured to communicate with the terminals 111-113 by wire or wireless.

The server 100 may work with a web page or application 101 . The webpage or application 101 may be a platform for introducing contests and managing careers of contests. The web page or application 101 is a contest participant or award winner; A person or organization that organizes or acts on behalf of the contest; The public works or award-winning works can be mainly used by persons or organizations who wish to view them for commercial use or other purposes.

The web page or application 101 may receive input of various types of contest information (eg, contest name, period, poster, and other contents). The web page or the main page of the application 101 may display contest holding information. Information required by the user is exposed at the top through the interest category algorithm, so that the user can quickly check the contest information list of interest. The web page or application 101 may include a community in which users can be active.

The web page or application 101 may include a portfolio management service. Contest participants or award winners may enter their contest entry or award information into the web page or application 101 . The server 100 may record contest entry or award information that is determined not to be plagiarism and has been authenticated by a person or organization supervising or acting on behalf of the contest in the blockchain. Information on public works or collected works recorded on the blockchain can be displayed in the portfolio of contest participants or those who have won awards. The server 100 may display only works verified through a verification process on the web page or application 101 in order to prevent illegal copying and use cases. In addition, the server 100 may determine whether or not plagiarism of contest entries or award-winning works may be performed, and only works determined to be non-plagiarism may be displayed on the web page or application 101.

Public offerings or collections displayed in the portfolio can be used for monetization. A person or organization who wishes to view the contested works or award-winning works for commercial use or other purposes can view the portfolio and check the contest entry or award information. The web page or application 101 may include functions such as UI/button/message/payment/online contract for realizing the sale and use of contest entries or prize-winning works. When a contest entry work or award-winning work is sold, the server 100 may record the monetization contents of the contest entry work or award-winning work in the blockchain.

Terminals 111-113 may be desktop computers, notebooks, tablets, smart phones, and the like. For example, as shown in FIG. 1 , the first terminal 111 may be a desktop; The second terminal 112 may be a notebook; The third terminal 113 may be a smart phone. Types of the terminals 111 to 113 may vary according to embodiments. The terminals 111 to 113 include an arithmetic function that a typical computer has; save/reference function; input/output function; And it may be configured to perform all or part of the control function. The terminals 111 to 113 may be configured to communicate with the server 100 by wire or wireless. The terminals 111 to 113 may access a web page linked with the server 100 or may have an application linked with the server 100 installed thereon. The terminals 111 to 113 may exchange data with the server 100 through a web page or application 101 .

Users 121-123 refer to accounts used by people or organizations. The first user 121 may be an account used by a participant in a contest or a person who has won an award. The second user 122 may be an account of a person or organization that organizes or acts on behalf of the contest. The third user 123 may be an account of a person or organization who wishes to view the contested works or award-winning works for commercial use or other purposes.

Users 121-123 may log in to terminals 111-113. For example, you can log in to the first terminal 111; The second user 122 can log in to the second terminal 112; The third user 123 may log in to the third terminal 113 . Users 121 to 123 logged into the terminals 111 to 113 may exchange data with the server 100 through a web page or application 101 . Each of the users 121 , 122 , and 123 has an authority to access each user information stored in the server 100 .

Meanwhile, for convenience of description, only three terminals 111 to 113 are shown in FIG. 1 , but the number may vary. As long as the processing capacity of the server 100 allows, the number of terminals is not particularly limited.

The database 130 may be a database owned by a person or organization that provides services using the server 100; It may be an external database linked with the server 100 through an API; It may be a set of distributed data stored in distributed nodes. The database 130 may be a relational database, an unstructured database (NoSQL), or a simple file storage. The database 130 may be managed by a general database management system (DBMS). The database 130 may be configured to communicate with the server 100 by wire or wireless. The database 130 may not be a single physical device, and may employ several physical devices according to functions, purposes, necessity, and retained data.

The database 130 may store contest entries or prize-winning works uploaded by users using the web page or application 101 . Contest entries or prize-winning works in the database 130 may be used for artificial intelligence learning. Artificial intelligence can be used in the process of determining plagiarism of contest entries or prize-winning works.

Hereinafter, embodiments will be described focusing on the operation of the server 100, and various application examples may be employed without being limited by the subject or aspect of communication. Referring to Figure 2, an embodiment relating to the operation of the system is described. Referring to FIG. 3, an embodiment related to a plagiarism determination operation will be described. Referring to FIG. 4 , an embodiment related to an operation of determining plagiarism according to a request of a contest organizer or agency will be described. An embodiment related to a data augmentation operation for learning an artificial neural network will be described with reference to FIG. 5 . Referring to FIG. 6 , an embodiment related to a learning operation of artificial intelligence will be described.

2 is a diagram for explaining the operation of a system according to an embodiment.

First, the server 100 may receive contest entry or award information from the first user 121 (210).

The first user 121 may be an account used by a participant in a contest or a person who has won an award. The first user 121 may input contest entry or award information into the web page or application 101 linked with the server 100 . Contest entry or award information may include contest entries or award-winning works, whether or not to use the contest entries or award-winning works for monetization, sales information on contest entries or award-winning works, and the like. Sales information may include a sales price.

The web page or application 101 may include a portfolio management service. The portfolio of the first user 121 may include introductions of contested works, user profiles, external links, introductions of similar works, and the like. The web page or application 101 may display information about contests in which contest participants or winners participated. The first user 121 may manage various portfolios of contest participation or award information like a photo album (gallery). The web page or application 101 may include an automatic arrangement and sorting function of contest entries or prize-winning works. In addition, it may include a detailed search of contest entries or prize-winning works, and functions to disclose/undisclose works. The third user 123 may select the portfolio of the first user 121 from the award/career inquiry page to check the career information of the first user 121 and the contest entry or prize-winning works.

Next, the server 100 may receive authentication for contest entry or award information from the second user 122 (220).

Authentication of contest entry or award information may be performed by the second user 122 . The second user 122 may be an account of a person or organization that organizes or acts on behalf of the contest. The second user 122 may input authentication for contest entry or award information into the web page or application 101 linked with the server 100 . The server 100 may receive authentication for contest entry or award information.

Authentication of the contest entry or award information may include verifying whether the contest entry or award-winning work has really been entered into or won a contest organized by or on behalf of the second user 122 . In addition, the certification of entry into the contest or award information may include whether or not the second user 122 permits the monetization of the award-winning work by the first user 121 .

In addition, the second user 122 may process all functions for the administration of the contest in the web page or application 101 linked with the server 100 . When the contest ends and the results are announced, the second user 122 may register the result information in the web page or application 101 linked with the server 100, which may be linked to the portfolio of the first user 121. there is.

Meanwhile, the second user 122 may not authenticate contest entry or award information when there is forgery or alteration. In this case, the server 100 may impose a predefined penalty on the first user 121 without recording contest entry or award information in the blockchain.

Subsequently, the server 100 may determine whether or not plagiarism of the contest entries or award-winning works (230).

The server 100 may include pre-learned artificial intelligence. The artificial intelligence may output an embedding vector of the contest entry or prize-winning work as input. The dimension of the embedding vector may be significantly lower than the original data dimension of the contest entries or award-winning works. The learning process of artificial intelligence will be described later with reference to FIG. 6 .

The server 100 may pre-store each embedding vector of each contest entry or award-winning work stored in the database 130 by applying artificial intelligence to the contest entries or award-winning works stored in the database 130 in advance. The server 100 may acquire an embedding vector of the first user 121's contest entry or award-winning work by applying artificial intelligence to the contest entry or award-winning work of the first user 121 . Using contest entries or award-winning works expressed as embedding vectors, the server 100 performs a first search of the contest entry or award-winning works stored in the database 130 and the contest entry or award-winning works of the first user 121. Similarity and first plagiarism can be judged.

After determining the first similarity and the first plagiarism, the server 100 may transmit the first similarity and the first plagiarism to the first user 121 (240). In addition, the first user 121's contest entries or prize-winning works may be stored in the database 130 (250).

A more detailed operation of the server 100 will be described later with reference to FIG. 3 .

The server 100 may record the results of the contest determined to be not plagiarism and verified by the second user 122 in the blockchain, and issue a QR code for the contest result. QR codes are issued in the form of a hash code rather than a form of URL address, and can be interpreted through a dedicated App. The QR code may include information such as the name of the winner, details of awards received in the contest, date of issuance, and organization in charge of the contest.

Contest entry or award information determined to be non-plagiarized and verified by the second user 122 may be converted into data into a block chain described later and recorded in a closed block chain. The first user 121 can check the award certificate online and output it offline. A QR code is loaded on the printed certificate of award and can be viewed through mobile devices. It is possible to fundamentally block forgery and alteration of certificates by comparing the output of the award certificate and the entry through the QR code.

In this way, the server 100 checks the authenticity of the contest award history of the first user 121 and plagiarism of contest entries or award-winning works, issues a reliable mark by using blockchain-based technology, and provides information on contest awards. reliability can be verified. When authentication for contest entry or award information is completed, an authentication QR code, verification mark, etc. may be issued, and the first user 121 may output corresponding award information.

On the other hand, the server 100 may determine the contest participation work or award-winning work as plagiarism. In this case, the server 100 may not record contest entry or award information in the blockchain, and may grant the first user 121 an opportunity to explain or impose a predefined penalty.

In the following order, the server 100 may record contest entry or award information in the blockchain (260).

The server 100 determines that the contest entries or award-winning works are not plagiarism, and that they have indeed been submitted to the contest organized or run by the second user 122 or won in the contest hosted or run by the second user 122. Only when authenticated, contest entry or award information can be recorded on the blockchain.

In the server 100, a Hyperledger Fabric blockchain platform architecture may be built on a private cloud system built with OpenStack. By using a private blockchain, blocks containing sensitive information can be prevented from being indiscriminately exposed to the outside world. However, the blockchain platform architecture is not limited thereto and may be employed differently depending on the embodiment.

The server 100 may determine whether the first user 121's contest entry work or award-winning work is plagiarized by inputting the contest entry work or award-winning work of the first user 121 to the pre-learned artificial intelligence. In addition, the server 100 may receive authentication of the contest entry or award information of the first user 121 from the second user 122 . Through this, it is possible to complete the verification of plagiarism of contest entries or award-winning works and whether the contest entry works or award-winning works have really been submitted to or won awards from the contest supervised or run by the second user 122 .

The server 100 judges that the contest entry or award-winning work is not plagiarism, and the contest entry or award-winning work is truly submitted to the contest hosted or won by the second user 122. Alternatively, award information can be recorded on a closed blockchain. Contest entry or award information may include contest entries or award-winning works, whether or not to use the contest entries or award-winning works for monetization, sales information on contest entries or award-winning works, and the like. Since all blockchain data change histories for each user are stored, it can be easy to track the history for each user.

On the other hand, the server 100 may determine the contest participation work or award-winning work as plagiarism. In addition, the second user 122 may not authenticate contest entry or award information when there is forgery or alteration. In this case, the server 100 may not record contest entry or award information in the blockchain, and may grant the first user 121 an opportunity to explain or impose a predefined penalty.

Blockchain can be recorded on a distributed ledger. According to one embodiment, the stored ledger information is collected in world-state-db, and can be inquired in the form of a JSON document through an API. World-state-db can store and track the ledger history of blocks. The integrity of the distributed ledger can be guaranteed by sharing the distributed ledger through real-time linkage between nodes. When a failure occurs in a stored node, synchronization and recovery of blocks can be performed automatically after failure recovery by agreement between nodes.

The first user 121 and the second user 122 can manage the contest entry or award information recorded in the blockchain through a web page or application 101 linked to the server 100 and search it with a QR code. . The first user 121 may inquire with a pre-issued QR code and output a confirmation when proving personal contest experience. The third user 123 may view contest entry or award information that is allowed to be viewed on the web page or application 101 . When searching for contest entry or award information, various conditions such as date and contest category may be specified.

In addition, the server 100 may use only minimal personal information through DID to authenticate awards. The server 100 may verify the identities of the second user 122 who is the organizer or agency of the contest and the first user 121 who is the contest participant through the DID module. When contest participants sign up or participate in the contest, they can go through the DID document registration process and store it in the DID-only blockchain. Integrity and denial can be prevented by the impossibility of tampering and withdrawal of the contents recorded in the blockchain. The issued DID document can be used for contest participant identification and contest participation certification. In addition, it can be used as an authentication means such as logging in on a web page or application 101 and participating in a contest.

When adding a new node, the server 100 may share data with an existing node. This can be done by adding a new node identical to the existing blockchain. The server 100 may synchronize with previous blocks of existing nodes as soon as a new node is added. The ledger of the existing node and the new node can be synchronized through the chaincode menu of the blockchain manager.

On the other hand, the server 100 may search contest entry works or prize-winning works stored in the database 130 during artificial intelligence learning (270).

The artificial intelligence included in the server 100 may receive a contest entry or award-winning work and output an embedding vector of the contest entry or award-winning work to be verified. The dimension of the embedding vector may be significantly lower than the original data dimension of the contest entries or award-winning works. For AI learning, the server 100 may use competition participation works or winning works stored in the database 130 as training data. The learning process of artificial intelligence will be described later with reference to FIG. 6 .

The server 100 may allow the third user 123 to view at least a portion of the blockchain record.

The web page or application 101 linked with the server 100 may display contest entry or award information in the portfolio of the first user 121 . The third user 123 can view the portfolio of the first user 121 on the web page or application 101 and check contest entry or award information. The third user 123 may be an account of a person or organization who wishes to view the contested works or award-winning works for commercial use or other purposes.

The third user 123 may search/search/view works participating in the contest or award-winning works for which sales and sharing functions are activated among the portfolios of the first user 121 . The web page or application 101 provides a function for obtaining permission to use the contest entry or award-winning work at the selling price suggested by the first user 121, and the use of the contest entry or award-winning work for the first user 121. UI/button/message/approval for realizing the sale and use of contest entries or award-winning works, such as a function of sending a message that the user 121 wants to do so, and a function of bidding the price of the contest entry or award-winning work to the first user 121 through an auction. / Can include functions such as online contracts. The third user 123 who has obtained the commercial right for the contest entry work or award-winning work through purchase completion may include a function of downloading the contest participation work or award-winning work purchased from the first user 121 .

On the other hand, when sales of contest entries or award-winning works are made, the server 100 may record sales records of contest entries or award-winning works in the blockchain. Records of sales of contest entries or prize-winning works include the name of the contest, organizer, agency, implementation year, whether the sold work is an entry or prize-winner, information on contest participants and winners, and the person who obtained commercial rights for the contest entry or award-winning work; It may include information about the organization, the sale price of contest entries or award-winning works, and the scope of use of contest entries or award-winning works.

In addition, the server 100 may allow the third user 123 to view at least a portion of the blockchain record.

The web page or application 101 linked with the server 100 may display contest entry or award information in the portfolio of the first user 121 . The third user 123 can view the portfolio of the first user 121 on the web page or application 101 and check contest entry or award information. The third user 123 may be an account of a person or organization who wishes to view the contested works or award-winning works for commercial use or other purposes.

The third user 123 may search/search/view works participating in the contest or award-winning works for which sales and sharing functions are activated among the portfolios of the first user 121 . The web page or application 101 provides a function for obtaining permission to use the contest entry or award-winning work at the selling price suggested by the first user 121, and the use of the contest entry or award-winning work for the first user 121. UI/button/message/approval for realizing the sale and use of contest entries or award-winning works, such as a function of sending a message that the user 121 wants to do so, and a function of bidding the price of the contest entry or award-winning work to the first user 121 through an auction. / Can include functions such as online contracts. The third user 123 who has obtained the commercial right for the contest entry work or award-winning work through purchase completion may include a function of downloading the contest participation work or award-winning work purchased from the first user 121 .

On the other hand, when sales of contest entries or award-winning works are made, the server 100 may record sales records of contest entries or award-winning works in the blockchain. Records of sales of contest entries or prize-winning works include the name of the contest, organizer, agency, implementation year, whether the sold work is an entry or prize-winner, information on contest participants and winners, and the person who obtained commercial rights for the contest entry or award-winning work; It may include information about the organization, the sale price of contest entries or award-winning works, and the scope of use of contest entries or award-winning works.

Through the above, the server 100 may determine that plagiarism is not plagiarism and record the result of the contest that has been verified by the second user 122 in the blockchain. Accordingly, the server 100 can provide contest entry or award information with public confidence. Activities and award information for various contests are recorded based on blockchain technology and can be proven with public confidence. Even if an unspecified number of contest award information is forged, the problem of counterfeiting can be solved through blockchain. The first user 121 may utilize the contest participation or award records recorded in the blockchain as his or her portfolio.

In addition, the server 100 may provide safe and simple career authentication. Even when the certificate or plaque (trophy) of the winner of the contest is damaged or lost and it is impossible to prove the history of the award, or when the authenticity is not confirmed due to various reasons such as poor management by the host organization or change in person in charge, History can be easily authenticated by using the web page or application 101 .

Furthermore, the server 100 may provide plagiarism determination for contest entries or prize-winning works. When it is determined to be plagiarism, the server 100 does not record contest entry or award information in the blockchain, and may grant the first user 121 an opportunity to explain or impose a predefined penalty. Through this, the web page or application 101 can prevent the exposure of contest entries or award-winning works with plagiarism controversy.

3 is a flowchart illustrating a plagiarism determination operation according to an embodiment.

Specifically, operations 230 to 250 of determining whether a work entered into a contest or an award-winning work of the first user 121 described with reference to FIG. 2 may include the following.

First, the server 100 may obtain an embedding vector of the first user 121's contest entry or award-winning work by applying the contest entry or award-winning work of the first user 121 to the previously learned artificial intelligence. (310).

The server 100 may include pre-learned artificial intelligence. The artificial intelligence may output an embedding vector of the contest entry or prize-winning work as input. The dimension of the embedding vector may be significantly lower than the original data dimension of the contest entries or award-winning works. The learning process of artificial intelligence will be described later with reference to FIG. 6 .

The server 100 may obtain an embedding vector of the first user 121's contest entry or award-winning work by applying the contest entry or award-winning work of the first user 121 to artificial intelligence. The embedding vector of the contest entries or award-winning works of the first user 121 compressively expresses the original contest entries or prize-winner data expressed in images, videos, text, and the like.

Next, the server 100 may determine a first similarity between each contest entry or award-winning work stored in the database 130 and the first plagiarism (320).

To this end, the server 100 applies artificial intelligence to the contest entries or award-winning works stored in the database 130 in advance and stores each embedding vector of each contest entry or award-winning work stored in the database 130 in advance. can Each embedding vector of each contest entry or prize-winning work stored in the database 130 compressively expresses original contest entry or award-winning work data expressed as images, videos, text, and the like.

In addition, the embedding vectors may be vectors having the same size and different directions. Therefore, the more similar the directions of the two embedding vectors are, the more similar the two works (image, video, text, etc.) can be.

The first degree of similarity may refer to a degree of similarity between contest entry works or prize-winning works stored in the database 130 and contest entry works or award-winning works of the first user 121 . The server 100 calculates the dot product of the embedding vector of the contest entries or award-winning works of the first user 121 and the embedding vectors of the respective contest entries or award-winning works stored in the database 130, or the cosine value of the angle formed by the two vectors. may be calculated with the first similarity. Meanwhile, various vector similarity measurement methods may be employed according to embodiments.

The server 100 may determine whether the first plagiarism is “plagiarism” when the first similarity is greater than or equal to a first predefined threshold. The first threshold may be determined as an average value of first similarity values of plagiarized works. According to other embodiments, the first threshold may be determined in various ways.

Subsequently, the server 100 may transmit a first similarity exceeding a first threshold and a first plagiarism status to the first user 121 (330).

When the server 100 determines that a contest entry or award-winning work is plagiarized, it does not record the contest entry or award information in the blockchain, and may give the first user 121 an opportunity to explain or impose a predefined penalty. can The first user 121 receiving the first similarity exceeding the first threshold and the first plagiarism status may proceed with an explanatory procedure in the web page or application 101 . When the calling procedure is accepted, the server 100 may record contest entry or award information of the first user 121 in the blockchain.

In the following order, the server 100 may store the contest entries or prize-winning works of the first user 121 in the database 130 regardless of the first plagiarism result (340).

Regardless of plagiarism or not, the work uploaded by the first user 121 can be used as artificial intelligence learning data by storing the contest participation work or award-winning work of the first user 121 in the database 130 . Through this, artificial intelligence can be learned using more data. Through this, artificial intelligence can improve the ability to express works composed of images, videos, texts, etc. with embedding vectors. In particular, if the original work and the plagiarized work are stored with the same label, and artificial intelligence is trained using the label, the artificial intelligence outputs the embedding vector for the original work and the embedding vector for the plagiarized work more similarly. can be learned as much as possible.

4 is a flowchart illustrating an operation of determining plagiarism according to a request from a contest organizer or agency according to an embodiment.

There are cases in which a person or organization that organizes or acts on behalf of a contest wants to check whether or not there is plagiarism in the contest entries or prize-winning works. For example, the organization in charge of the contest may want to confirm whether the work is not plagiarized before awarding the contest entry subject to evaluation. In addition, the contest agency may want to increase the reliability of work by checking whether there are plagiarism in contest entries or prize-winning works managed by them and informing the contest host organization. In order to meet such demand, the server 100 may include a plagiarism determination operation according to a request from a contest organizer or agency.

First, the server 100 may receive a contest entry work or award-winning work to be verified from the second user 122 (410).

The second user 122 may be an account of a person or organization that organizes or acts on behalf of the contest. The works participating in the contest or winning works subject to verification may be contested works that the organization in charge of the contest wants to check whether or not the works are plagiarized prior to awarding the contested works to be judged. Alternatively, it may be a contest entry or prize-winning work managed by a contest agency. There may be multiple entries or award-winning works in the contest subject to verification.

Next, the server 100 may acquire an embedding vector of a contest entry or award-winning work subject to verification by applying the contest entry or award-winning work to be verified (420).

The server 100 may include pre-learned artificial intelligence. The artificial intelligence can output the embedding vector of the contest entry or award-winning work to be verified by receiving the contest entry or award-winning work. The dimension of the embedding vector may be significantly lower than the original data dimension of the contest entries or award-winning works. The learning process of artificial intelligence will be described later with reference to FIG. 6 .

The server 100 may obtain an embedding vector of a contest entry or award-winning work to be verified by applying the contest entry or award-winning work to be verified to artificial intelligence. The embedding vector of the contest entry or award-winning work to be verified compressively expresses the original contest entry or award-winning data expressed in images, videos, text, etc.

Subsequently, the server 100 may determine a second similarity between each contest entry or award-winning work stored in the database and a contest entry or award-winning work to be verified (430).

As described with reference to FIG. 3 , the server 100 pre-applies the contest entries or award-winning works stored in the database 130 to artificial intelligence, and each embedding of each contest entry or award-winning work stored in the database 130. Vectors can be stored in advance. Each embedding vector of each contest entry or prize-winning work stored in the database 130 compressively expresses original contest entry or award-winning work data expressed as images, videos, text, and the like.

Embedding vectors may be vectors having the same magnitude and different directions. Therefore, the more similar the directions of the two embedding vectors are, the more similar the two works (image, video, text, etc.) can be.

The second degree of similarity may refer to a similarity between each contest entry or award-winning work stored in the database 130 and the contest entry or award-winning work subject to verification. The server 100 converts the dot product of the embedding vector of the contest entry or award-winning work to be verified and the embedding vector of each contest entry or award-winning work stored in the database 130 or the cosine value of the angle formed by the two vectors to the second similarity degree. can be computed. Meanwhile, various vector similarity measurement methods may be employed according to embodiments.

The server 100 may determine whether the second plagiarism is "plagiarism" when the second similarity is equal to or greater than a predefined second threshold. The second threshold may be determined as an average value of second similarities of plagiarized works. According to other embodiments, the second threshold may be determined in various ways.

In the following order, the server 100 may transmit to the second user 122 a second similarity exceeding a second threshold and a second plagiarism status (440).

If the original work is open to the public, the server 100 informs the second user 122 of the second similarity exceeding the second threshold and the second plagiarism, as well as the suspected plagiarism of the contest entry or award-winning work subject to verification. Original works can be provided. The second user 122 refers to the second similarity and whether or not the second plagiarism is used to determine whether the contest participation works or award-winning works subject to verification are awarded, whether the award-winning works are revoked, whether or not the contest participants are penalized, etc. can decide

5 is a diagram for explaining a data augmentation operation for learning an artificial neural network according to an embodiment.

The server 100 may include artificial intelligence. Artificial intelligence may be learned through the server 100 or a separate learning device. For artificial intelligence artificial neural network learning, the server 100 or the learning device may include a data aggregation module 500 .

The data aggregation module 500 may generate inputs and labels to be applied to artificial neural network learning by repeatedly performing pre-processing of transforming contest entries or prize-winning works stored in the database 130 . For example, the contest winning work 510 may be an image. The data aggregation module 500 may include conventional modules for image editing. The data aggregation module (500) is a version of the enlargement (511), reduction (512), color change (513), filter effect (514), rotation (515), and inversion (516) version of the contest prize-winning work (510). images can be created. Various versions of images combining the image editing effects 511 to 516 may be created, and various versions of images not shown may also be created.

Images generated through the data aggregation module 500 may be viewed as plagiarized works of the prize-winning works 510 of the contest. The server 100 or the learning device may use the original competition-winning work 510 and plagiarized images generated through the data aggregation module 500 as input data for artificial neural network learning. At this time, the server 100 or the learning device may assign the same label to the plagiarism images generated through the original competition winning work 510 and the data aggregation module 500 .

Through a process described below with reference to FIG. 6 , the server 100 or the learning device may train the artificial neural network so that the artificial neural network outputs embedding vectors of input data having the same label as similarly as possible. Accordingly, the embedding vector of the plagiarized work output by the artificial intelligence that has completed learning may be similar to the embedding vector of the original work.

On the other hand, although the case where the contest participation work or award-winning work is an image has been described above, the data aggregation module 500 is not limited thereto. The data aggregation module 500 may have a typical configuration capable of performing preprocessing on data composed of images, texts, and the like, according to embodiments.

6 is a diagram for explaining learning of artificial intelligence according to an embodiment.

Artificial intelligence may be a component included in the server 100 and may be learned through the server 100 or a separate learning device.

The artificial intelligence may be trained to receive contest entries or prize-winning works as input and output embedding vectors of the contest entries or award-winning works. It should be noted that an artificial neural network in the form of an encoder-decoder is trained to train an artificial intelligence model that generates an embedding vector, but only the weight of the encoder unit is used after learning is completed. That is, the server 100 uses artificial intelligence that partially uses only weights corresponding to the encoder unit among the total weights of the artificial neural network learned by the learning device.

Hereinafter, a process in which artificial intelligence is learned through the learning device will be described.

First, the learning device may acquire training data (600).

For training of the entire artificial neural network composed of encoders and decoders, the learning device may acquire contest entries or prize-winning works stored in the database 130 .

Next, the learning device may generate an input and a label of the artificial neural network from the training data (610).

The learning device obtains various versions of plagiarized works for each contest entry or award-winning work by applying training data, each contest participation work or award-winning work, to the data aggregation module 500 described with reference to FIG. can The learning device may use original contest entries or prize-winning works and plagiarism contest entries or award-winning works generated through the data aggregation module 500 as input data for artificial neural network learning. In this case, the learning device may assign the same label to the original contest entries or prize-winning works and plagiarism contest entries or award-winning works generated through the data aggregation module 500 .

The learning device may use a stochastic gradient descent method in the artificial intelligence optimization step 650 for learning efficiency, and for this purpose, the artificial intelligence input data may be distributed by batch. Training data may be redistributed within a batch for each learning epoch.

Next, the learning device may apply the input to artificial intelligence (620).

The entire artificial neural network to be trained may be in the form of an encoder-decoder. Since the contest entries or award-winning works are images, videos, texts, etc., detailed modules of artificial neural networks may include convolutional neural networks (CNNs), recurrent neural networks (RNNs), transformers, and combinations thereof.

Subsequently, the learning device may obtain an output from artificial intelligence (630).

The encoder unit may receive the original contest entries or prize-winning works and plagiarism contest entries or award-winning works generated through the data aggregation module 500 and output embedding vectors for them. The decoder unit may receive an output of the encoder unit and output an estimation of the label.

Then, the learning device may compare the output and the label (640).

The process of comparing the output of the artificial neural network corresponding to inference and the label corresponding to the correct answer may be performed by calculating a loss function. As the loss function, a known mean squared error (MSE), cross entropy error (CEE), or the like may be used. However, it is not limited thereto, and loss functions used for various artificial intelligence models can be used if the deviation, error or difference between the output of the artificial neural network and the label can be measured.

Next, the learning device may optimize artificial intelligence based on the comparison value (650).

By updating the weights of the nodes of the artificial neural network model so that the learning device comparison value gradually decreases, the output of artificial intelligence corresponding to inference and the label corresponding to the correct answer can be gradually matched, and through this, artificial intelligence can be optimized to output an inference close to the correct answer. Specifically, the learning device may optimize artificial intelligence by repeating a process of resetting a weight of artificial intelligence so that a loss function corresponding to a comparison value approaches an estimated minimum value. For optimization of artificial intelligence, a known backpropagation algorithm, stochastic gradient descent, and the like may be used. However, it is not limited thereto, and weight optimization algorithms used in various neural network models may be used.

The learning device may learn the decoder unit by gradually matching the output of the decoder unit and the label. At the same time, the encoder part can be taught to produce an output that allows the output of the decoder part to match the label as closely as possible. When learning of the artificial neural network is completed, the encoder unit and the decoder unit may be optimized.

The server 100 may include artificial intelligence using weights of the encoder unit. The artificial intelligence included in the server 100 may generate an embedding vector by receiving a contest entry or award-winning work. The dimension of the embedding vector may be significantly lower than the original data dimension of the contest entries or award-winning works.

7 is an exemplary diagram of a configuration of a device according to an embodiment.

Device 701 according to an embodiment includes a processor 702 and a memory 703 . The device 701 may include at least one of the devices described above with reference to FIGS. 1 to 6 or may perform at least one operation described above with reference to FIGS. 1 to 6 . Specifically, the device 701 may be a server 100, a first terminal 111, a second terminal 112, a third terminal 113, a database 130, or an artificial intelligence learning device. A person or organization using the device 701 may provide a service related to some or all of the methods described above with reference to FIGS. 1 to 6 .

The memory 703 may store information related to the above-described methods or a program in which the above-described methods are implemented. Memory 703 may be volatile memory or non-volatile memory.

The processor 702 may execute a program and control the device 701 . Codes of programs executed by the processor 702 may be stored in the memory 703 . The device 701 may be connected to an external device (eg, a personal computer or network) through an input/output device (not shown) and exchange data through wired/wireless communication.

The device 701 may be used to train artificial intelligence or to use learned artificial intelligence. The memory 703 may contain artificial intelligence that is being learned or has been learned. The processor 702 may learn or execute an artificial intelligence algorithm stored in the memory 703 . The device 701 for learning artificial intelligence and the device 701 for using the learned artificial intelligence may be the same or separate.

The embodiments described above may be implemented as hardware components, software components, or a combination of hardware components and software components. For example, the devices, methods and components described in the embodiments may include, for example, a processor, a controller, an arithmetic logic unit (ALU), a digital signal processor, a microcomputer, a field programmable gate (FPGA). array), programmable logic units (PLUs), microprocessors, or any other device capable of executing and responding to instructions. The processing device may run an operating system (OS) and one or more software applications running on the operating system. A processing device may also access, store, manipulate, process, and generate data in response to execution of software. For convenience of understanding, there are cases in which one processing device is used, but those skilled in the art will understand that the processing device includes a plurality of processing elements or a plurality of types of processing elements. know you can do it. For example, a processing device may include a plurality of processors or a processor and a controller. Other processing configurations are also possible, such as parallel processors.

The method according to the embodiment may be implemented in the form of program instructions that can be executed through various computer means and recorded on a computer readable medium. Computer readable media may include program instructions, data files, data structures, etc. alone or in combination. Program commands recorded on the medium may be specially designed and configured for the embodiment or may be known and usable to those skilled in computer software. Examples of computer-readable recording media include magnetic media such as hard disks, floppy disks and magnetic tapes, optical media such as CD-ROMs and DVDs, and magnetic media such as floptical disks. - includes hardware devices specially configured to store and execute program instructions, such as magneto-optical media, and ROM, RAM, flash memory, and the like. Examples of program instructions include high-level language codes that can be executed by a computer using an interpreter, as well as machine language codes such as those produced by a compiler. A hardware device may be configured to act as one or more software modules to perform the operations of the embodiments, and vice versa.

Software may include a computer program, code, instructions, or a combination of one or more of the foregoing, which configures a processing device to operate as desired or processes independently or collectively. The device can be commanded. Software or data may be any tangible machine, component, physical device, virtual equipment, computer storage medium or device, or It can be permanently or temporarily embodied in a transmitted signal wave. Software may be distributed on networked computer systems and stored or executed in a distributed manner. Software and data may be stored on one or more computer readable media.

As described above, although the embodiments have been described with limited drawings, those skilled in the art can apply various technical modifications and variations based on this. For example, the described techniques may be performed in a different order than the described method, or components of the described system, structure, device, circuit, etc. may be combined or combined in a different form than the described method, or other components or equivalents may be used. Appropriate results can be achieved even when substituted or substituted by water.

Therefore, other implementations, other embodiments, and equivalents of the claims are within the scope of the following claims.

100: server
101: Web page or application
111, 112, 113: first to third terminals
121, 122, 123: first to third users
130: database

Claims (5)

A method for managing entries for a blockchain-based contest performed by at least one server,
Receiving contest entry or award information from a first user; and
Receiving authentication for the contest entry or award information from a second user
including,
The above contest entry or award information includes contest entries or award-winning works,
Authentication of the contest entry or award information includes verification of whether the contest entry or award-winning work was really entered into the contest organized or run by the second user or won in the contest hosted or run by the second user. do,
The blockchain-based contest entry management method
Determining whether the contest entries or award-winning works are plagiarized; and
Recording the contest entry or award information in the blockchain
Including more,
Only when it is judged that the work participating in the contest or the winning work is not plagiarism, and it has been truly submitted to the contest organized or run by the second user, or has been certified as an award winner in the contest organized or run by the second user, entry into the contest Or record the award information in the blockchain,
The server is linked with a database distinct from the blockchain,
The step of determining whether the contest entries or award-winning works are plagiarized,
Determining a first similarity between each contest entry or award-winning work stored in the database and first plagiarism; and
Storing the contest entries or award-winning works in the database, regardless of the first plagiarism result.
including,
The server includes artificial intelligence,
Contest entries or award-winning works stored in the database are used for learning of the artificial intelligence,
Blockchain-based contest entry management method.
According to claim 1,
The step of determining the first similarity between each contest entry or award-winning work stored in the database and the first plagiarism,
Acquiring each embedding vector of each contest entry or award-winning work stored in the database by applying the artificial intelligence to each contest entry or award-winning work stored in the database.
containing
Blockchain-based contest entry management method.
According to claim 2,
The step of determining the first similarity between each contest entry or award-winning work stored in the database and the first plagiarism,
obtaining an embedding vector of the contest entry or award-winning work by applying the contest entry or award-winning work to the artificial intelligence;
Determining the first similarity and first plagiarism based on the similarity between each embedding vector of each contest entry or prize-winning work stored in the database and the embedding vector of the contest entry or award-winning work; and
Transmitting each first similarity exceeding a first threshold and a first plagiarism status to the first user
containing
Blockchain-based contest entry management method.
According to claim 2,
receiving a contest entry or award-winning work to be verified from the second user;
obtaining an embedding vector of the contest entry or award-winning work to be verified by applying the contest entry or award-winning work to the artificial intelligence;
determining a second degree of similarity between each contest entry or prize-winning work stored in the database and a second plagiarism; and
Transmitting the second similarity exceeding a second threshold and the second plagiarism status to the second user
Including more,
Blockchain-based contest entry management method.
Blockchain-based contest entry management device
Receive contest entry or award information from the first user;
Receiving authentication for the contest entry or award information from the second user
including at least one processor that performs an operation;
The above contest entry or award information includes contest entries or award-winning works,
Authentication of the contest entry or award information includes verification of whether the contest entry or award-winning work was really entered into the contest organized or run by the second user or won in the contest hosted or run by the second user. do,
The processor
Judging whether the works participating in the contest or winning works are plagiarized,
Record the contest entry or award information in the blockchain,
Only when it is judged that the work participating in the contest or the winning work is not plagiarism, and it has been truly submitted to the contest organized or run by the second user, or has been certified as an award winner in the contest organized or run by the second user, entry into the contest Or record the award information in the blockchain,
The device is linked to a database distinct from the blockchain,
The operation of the processor to determine whether the competition participation work or award-winning work is plagiarized,
Determining the first similarity and the first plagiarism between each contest entry or award-winning work stored in the database and the contest entry entry or award-winning work;
Regardless of the first plagiarism result, storing the contest participation work or award-winning work in the database
include action,
The device includes artificial intelligence,
Contest entries or award-winning works stored in the database are used for learning of the artificial intelligence,
Blockchain-based contest entry management device.

Images