WO2022191380A1

WO2022191380A1 - Blockchain-based image forgery and falsification preventing system and method, and computer program therefor

Info

Publication number: WO2022191380A1
Application number: PCT/KR2021/017935
Authority: WO
Inventors: 김명호; 권재환; 장성일
Original assignee: 숭실대학교 산학협력단
Priority date: 2021-03-10
Filing date: 2021-12-01
Publication date: 2022-09-15
Also published as: KR102516548B1; KR102516548B9; KR20220126932A

Abstract

A blockchain-based image forgery and falsification preventing system comprises: an input unit configured to receive image information from an imaging device; an image segmentation unit configured to divide the image information into a plurality of unit image data items; a hash extraction unit configured to calculate the respective hash value of the plurality of unit image data items; and a storage unit configured to record the respective hash value of the plurality of unit image data items in data blocks in a blockchain network, wherein each of the data blocks comprises the hash value of a preceding data block and the hash value of the unit image data, the storage unit recording the hash value of unit image data preceding the unit image data corresponding to each data block from among the plurality of unit image data items, in a data block preceding the respective data blocks. By using the image storage system, image data such as a video file is divided for each predetermined unit time (for example, frame) and the respective hash values of the divided parts are stored in the form of time-specific chains in data blocks in a blockchain network, and thus forgery and falsification, or fabrication of the image information may be prevented.

Description

Blockchain-based image forgery prevention system and method and computer program for the same

Embodiments relate to a blockchain-based image forgery prevention system and method, and a computer program for the same. More specifically, the embodiments relate to a technology for preventing forgery or manipulation of a stored image by applying a block chain technology to the storage of image information in a device that shoots and stores an image, such as a black box or CCTV.

Companies that have a lot of information that needs to be kept secure, such as trade secrets, install a control system on the door to control the access of outsiders to the facility or manage the access of users. CCTV cameras are often installed. In addition, CCTVs are installed for the purpose of crime prevention in enclosed spaces such as elevators or spaces frequented by many people.

For example, Patent Publication No. 10-2019-0135634 discloses a real-time crime monitoring elevator that can monitor incidents and accidents occurring inside the elevator by transmitting the image of CCTV installed inside the elevator to the monitors installed on each floor.

However, the image stored in the video recording device such as CCTV or black box can be easily seized or seized by a malicious third party, so the image may be forged or manipulated, and the original image may be easily damaged. there is a problem Because of these limitations, even when the video data is a major evidence for a crime or incident, the authenticity of the video cannot be trusted due to the possibility of forgery or manipulation by others, and thus the value as evidence decreases.

[Prior art literature]

[Patent Literature]

(Patent Document 1) Patent Publication No. 10-2019-0135634

According to one aspect of the present invention, by applying a blockchain-based technology to the storage of image information in a device that shoots and stores images, such as a black box and CCTV, it is possible to check forgery or manipulation of an image in advance, It is possible to provide a block chain-based image forgery prevention system and method that can strengthen the reliability of stored images so that it is possible to use the image data as evidence, and a computer program for this.

An image forgery prevention system based on a blockchain according to an embodiment includes an input unit configured to receive image information from a photographing device; an image dividing unit configured to divide the image information into a plurality of unit image data; a hash extraction unit configured to calculate a hash value of each of the plurality of unit image data; and a hash value of each of the plurality of unit image data is recorded in data blocks on a blockchain network, wherein each of the data blocks includes a hash value of a preceding data block and a hash value of the unit image data, and a storage unit configured to record, in a data block preceding each data block, a hash value of unit image data preceding the unit image data corresponding to each data block among the unit image data.

The block chain-based image forgery prevention system according to an embodiment uses images included in the image information as input values for a machine learning-based analysis model to detect objects corresponding to the images. It further includes a configured machine learning unit. In this case, the image dividing unit specifies an image in which at least one of a type of an object detected by the machine learning unit and the number of detected objects is changed, and converts before and after the specified image in the image information into different unit image data is further configured to divide.

In one embodiment, the storage unit stores transaction information on a token on the blockchain network defined corresponding to the unit image data to data blocks on the blockchain network together with the unit image data. further configured to record.

In one embodiment, the storage unit, based on the type or number of objects detected by the machine learning unit with respect to the image included in the unit image data, the transaction information or the entry/exit details of the token defined by the transaction information It is further configured to determine one or more of the quantity of tokens defined by .

The block chain-based image forgery prevention system according to an embodiment further includes a user management unit configured to receive user information corresponding to at least a part of the image information, including pre-registered user-specific wallet address information. In this case, the storage unit is further configured to determine a wallet address for the unit image data corresponding to each data block based on the user information, and to define the transaction information using the determined wallet address as a sender or a recipient.

In an embodiment, the storage unit is further configured to store the image information in the image storage system or an external storage.

A block-chain-based image forgery prevention method according to an embodiment includes: receiving, by a block-chain-based image forgery prevention system, image information from a photographing device; dividing, by the block chain-based image forgery prevention system, the image information into a plurality of unit image data; calculating, by the block chain-based image forgery prevention system, a hash value of each of the plurality of unit image data; and recording, by the blockchain-based image forgery prevention system, a hash value of each of the plurality of unit image data in data blocks on a blockchain network. In this case, each of the data blocks includes a hash value of the preceding data block and a hash value of the unit image data. In addition, the step of writing to the data blocks on the block chain network includes, by the block chain-based image forgery prevention system, a unit preceding the unit image data corresponding to each data block among the plurality of unit image data. and writing a hash value of the image data in a data block preceding each data block.

In the blockchain-based image forgery prevention method according to an embodiment, before the step of dividing the plurality of unit image data, the blockchain-based image forgery prevention system converts the images included in the image information into a machine learning-based The method further includes detecting an object corresponding to the images by using it as an input value for the analysis model. At this time, in the step of dividing the plurality of unit image data, the block chain-based image forgery prevention system changes at least one of the types of objects detected by the machine learning-based analysis model and the number of detected objects specifying an image to be used; and dividing, by the block chain-based image forgery prevention system, before and after the specified image in the image information into different unit image data.

In the block chain-based image forgery prevention method according to an embodiment, the block chain-based image forgery prevention system transmits transaction information for a token on the block chain network defined corresponding to the unit image data to the unit image data. and writing data blocks on the blockchain network together.

In one embodiment, the step of recording the transaction information in the data blocks on the block chain network includes, by the block chain-based image forgery prevention system, the machine learning-based analysis of the image included in the unit image data. and determining, based on the type or number of objects detected by the model, at least one of a token entry/exit details defined by the transaction information or a quantity of tokens defined by the transaction information.

A block-chain-based image forgery prevention method according to an embodiment comprises: storing wallet address information for each user registered in advance in the block-chain-based image forgery prevention system; and receiving, by the block chain-based image forgery prevention system, user information corresponding to at least a portion of the image information. At this time, the step of recording the transaction information in the data blocks on the block chain network includes, by the block chain-based image forgery prevention system, for the unit image data corresponding to each data block based on the user information. determining a wallet address; and defining, by the block chain-based image forgery prevention system, the transaction information using the determined wallet address as a sender or a recipient.

The block-chain-based image forgery prevention method according to an embodiment further includes, by the block-chain-based image forgery prevention system, storing the image information in the image storage system or an external storage.

The computer program according to an embodiment is to be combined with hardware to execute the block chain-based image forgery prevention method according to the above-described embodiments, and may be stored in a computer-readable medium.

Using a blockchain-based image forgery prevention system and method according to an aspect of the present invention, image data such as a moving picture file is divided by a predetermined unit time (eg, frame) and hashed. A value can be generated and these hash values can be stored in a time-wise chain in data blocks on the blockchain network. According to the image storage system and method, if there is forgery or manipulation of the image data, the hash value of the image data is different from the hash value stored on the block chain, so that it can be recognized that the image has been manipulated. . As a result, it is possible to prevent forgery of image data, and when the image is used as evidence for illegal activities such as crimes or incidents, there is an advantage in that reliability can be increased.

1 is a schematic block diagram of an image forgery prevention system based on a blockchain according to an embodiment.

2 is a flowchart illustrating each step of a method for preventing forgery and falsification of an image based on a block chain according to an embodiment.

3 is a conceptual diagram of a process of segmenting unit image data by applying object detection by machine learning in a block chain-based image forgery prevention method according to an embodiment.

4 is a conceptual diagram of a process of storing a hash value of image data in a data block on a block chain in a block chain-based image forgery prevention method according to an embodiment.

5 is a conceptual diagram for explaining video monitoring through a transaction of a block chain-based token in a block chain-based video forgery prevention method according to an embodiment.

Hereinafter, embodiments of the present invention will be described in detail with reference to the drawings.

Referring to FIG. 1 , the block chain-based image forgery prevention system 3 according to the present embodiment receives image information from one or more photographing devices 1 and blocks information necessary to prevent forgery of the received image information. It serves to store in the distributed ledger on the chain network (2). For example, the photographing device 1 may be a black box or CCTV, but is not limited thereto. Also, in one embodiment, the block chain-based image forgery prevention system 3 may store the image information itself in the internal storage of the block chain-based image forgery prevention system 3 or the storage of the external server 4 .

For the above operation, the block chain-based image forgery prevention system 3 is configured to be communicatively connected with the block chain network 2 and/or the external server 4 through a wired and/or wireless network. Also, in an embodiment, the block chain-based image forgery prevention system 3 may receive image information from one or more photographing devices 1 in a communication method through a network. In this specification, the communication method between the block chain-based image forgery prevention system 3 and the photographing device 1, the block chain network 2 and/or the external server 4 is an object through a wired and/or wireless network. It may include all communication methods that the and object can network, and is not limited to wired communication, wireless communication, 3G, 4G, or other methods.

For example, the blockchain-based image forgery prevention system (3) is LAN (Local Area Network), MAN (Metropolitan Area Network), GSM (Global System for Mobile Network), EDGE (Enhanced Data GSM Environment), HSDPA (High Speed Downlink Packet Access), W-CDMA (Wideband Code Division Multiple Access), CDMA (Code Division Multiple Access), TDMA (Time Division Multiple Access), Bluetooth, Zigbee, Wi-Fi ), Voice over Internet Protocol (VoIP), LTE Advanced, IEEE802.16m, WirelessMAN-Advanced, HSPA+, 3GPP Long Term Evolution (LTE), Mobile WiMAX (IEEE 802.16e), UMB (formerly EV-DO Rev. C), By at least one communication method selected from the group consisting of Flash-OFDM, iBurst and MBWA (IEEE 802.20) systems, HIPERMAN, Beam-Division Multiple Access (BDMA), Wi-MAX (World Interoperability for Microwave Access), and ultrasound-based communication. It may be configured to communicate, but is not limited thereto.

The blockchain network 2 includes a plurality of nodes 200, and each node 200 is communicatively connected to each other through a peer-to-peer network, The generated transaction is shared, and the node 200 determined in the mining competition by the consensus algorithm creates a data block to be recorded in the distributed ledger and stores the transaction information in the generated block. The chain of generated data blocks is shared by all nodes 200 of the blockchain network 2, and the nodes 200 synchronize the chain of blocks.

At this time, the block chain-based image forgery prevention system 3 according to the embodiments divides the image information received from the photographing device 1 into predetermined units to generate image data, and blocks the hash value of each image data. It functions to prevent forgery and falsification of image data by storing it in a data block on the chain network (2).

In one embodiment, the block chain-based image forgery prevention system 3 includes an input unit 31 , an image segmentation unit 32 , a hash extraction unit 34 , and a storage unit 36 . Also, in one embodiment, the block chain-based image forgery prevention system 3 further includes a machine learning unit 33 for intelligently performing segmentation of image information. Also, in one embodiment, the blockchain-based image forgery prevention system 3 further includes a storage 35 for storing the image information itself in the blockchain-based image forgery prevention system 3 . Further, in one embodiment, the block chain-based image forgery prevention system 3 further includes a user management unit 37 for reflecting the user in the information recorded in the block chain network 2 .

The block chain-based image forgery prevention system 3 and each part 31-37 included therein according to the embodiments may have an aspect that is entirely hardware, or is partially hardware and partially software. For example, the block chain-based image forgery prevention system 3 and each unit 31-37 included therein use hardware and related software to process data in a specific format and content and/or send/receive data in an electronic communication method. can be called As used herein, terms such as “unit”, “module”, “device”, “terminal”, “server” or “system” are intended to refer to a combination of hardware and software driven by the hardware. For example, the hardware may be a data processing device including a CPU or other processor. In addition, software driven by hardware may refer to a running process, an object, an executable file, a thread of execution, a program, and the like.

In addition, each element constituting the block chain-based image forgery prevention system 3 is not necessarily intended to refer to separate devices physically separated from each other. That is, each unit 31-37 of the block-chain-based image forgery prevention system 3 shown in FIG. 1 uses the hardware constituting the block-chain-based image forgery prevention system 3, an operation performed by the corresponding hardware. It is only functionally classified according to the requirements, and each part does not necessarily have to be provided independently of each other. Of course, depending on the embodiment, it is also possible to implement one or more of the above-described respective units 31 to 37 as separate devices physically separated from each other.

The input unit 31 may receive image information from the input unit 31 by being directly connected to one or more photographing devices 1 or by a communication method through a wired and/or wireless network. In the embodiments of the present specification, the blockchain-based image forgery prevention system 3 is shown as a separate block from the photographing device 1, but according to the embodiment, the blockchain-based image forgery prevention system 3 is photographed It may be implemented as a part of the imaging device 1 located in the device 1 . In addition, the number of photographing devices 1 shown in the drawing is merely exemplary, and does not limit the number of photographing devices 1 that operate in relation to the blockchain-based image forgery prevention system 3 .

The image dividing unit 32 may divide the image information received from the photographing apparatus 1 to the input unit 31 into a plurality of unit image data. For example, in an embodiment, the unit image data may refer to each frame of an image in image information having the same form as a moving picture file, but is not limited thereto, and a plurality of preset frames or Unit image data may be defined by using image information corresponding to a preset unit time.

In an embodiment, the image dividing unit 32 may determine a division into which unit image data is to be divided based on a result of detecting an object included in the image information by the machine learning unit 33 . For example, the machine learning unit 33 may detect an object included in the image by using the images included in the image information as input values to the machine learning-based analysis model. At this time, the image dividing unit 32 specifies, as a boundary image, an image in which the type and/or number of objects detected by the machine learning unit 33 is changed among a series of images constituting the image information, and the specified boundary image Front and back may be divided into different unit image data. In this case, each unit image data may have different lengths.

The hash extractor 34 may extract a hash value by applying a predetermined hash algorithm to each unit image data generated by the above-described process. Also, the storage unit 36 may store the hash value of each unit image data extracted by the hash extraction unit 34 in a data block on the blockchain network 2 . In the blockchain network 2, data blocks are sequentially generated, and each data block stores a hash value of the previous data block. In this case, the storage unit 36 may store the hash values of each continuous unit image data in time series in successive data blocks on the block chain, respectively.

As a result, the data blocks of the block chain each include the hash value of the unit image data together with the hash value of the preceding data block, and the hash values of the unit image data consecutive in time series are stored in the successive data blocks, respectively. do. That is, when contiguous first and second data blocks exist and contiguous first and second image data exist, a hash value of the first unit image data may be stored in the first data block, and the second The hash value of the first data block may be stored in the data block together with the hash value of the second unit image data. In this structure of data blocks in which the previous block stores the hash value, if some data block on the block chain is tampered with, the hash value to be recorded in the data blocks after it changes due to the characteristics of the hash function. It can be determined whether the image information has been manipulated through the change of the hash value.

The storage unit 36 may include a communication unit 361 for storing a hash value generated using a hash function in a data block on the blockchain network 2 . Also, the communication unit 361 may store unit image data or image information itself in the storage of the external server 4 through a wired and/or wireless network. Alternatively, in one embodiment, the storage unit 36 may be configured to store the unit image data or image information itself in the storage 35 of the block chain-based image forgery prevention system 3 .

In an embodiment, the storage unit 36 may further include a transaction generating unit 362 that generates transaction information to be recorded in a data block on the block chain together with a hash value of the unit image data. The transaction generator 362 may record transaction information about a token in which the input/output details are recorded in the distributed ledger defined by the data blocks on the block chain in the data block. In this case, the movement of the token defined by the transaction information may be determined based on the characteristic of the unit image data in which the hash value is recorded in the corresponding data block. For example, the transaction information may have the form of a smart contract, but is not limited thereto.

In an embodiment, the transaction generating unit 362 may define a token transaction according to the type and/or number of objects detected from the image by the machine learning unit 33 . Alternatively, in one embodiment, the block chain-based image forgery prevention system 3 receives user information related to image information through the user management unit 37, and the transaction generation unit 362 corresponds to the unit image data. A transaction can also be defined to reflect user information (eg, wallet address) in token movement. The form of a transaction reflecting the object or user information detected from the image will be described later in detail.

In one embodiment, the storage unit 36 records the image information itself in the internal storage of the external server 4 or the block chain-based image forgery prevention system 3, but the storage location of the image information in this storage can be specified. Possible location information may be further recorded in data blocks on the blockchain network 2 .

Also, in one embodiment, the storage unit 36 records a hash value generated from each unit image data of the image information in a data block on the blockchain network 2, but the image information itself is the main of the blockchain network 2 You can upload data blocks on the sidechain rather than the main net. When uploading video information to a sidechain, by introducing a sidechain that uses a consensus algorithm such as Proof of Authority (POA), the transaction processing speed can be greatly increased and the fee burden can be lowered. .

2 is a flowchart illustrating each step of a method for preventing forgery and falsification of an image based on a block chain according to an embodiment. The blockchain-based image forgery prevention method according to the embodiments may be performed using the blockchain-based image forgery prevention system according to the embodiments of the present invention. Hereinafter, for convenience of description, each step of the block chain-based image forgery prevention method according to the embodiments will be described with reference to FIGS. 1 and 2 .

1 and 2 , the input unit 31 of the block chain-based image forgery prevention system 3 may receive image information from one or more photographing devices 1 ( S1 ). As described above, the image dividing unit 32 may divide the image information into a plurality of unit image data, and each unit image data may have a preset number of frames or a preset time.

Meanwhile, in an embodiment, the image segmentation unit 32 may segment unit image data based on object recognition in the image by the machine learning unit 33 . In the present embodiment, the machine learning unit 33 may recognize an object in the image by using each image constituting the image information as an input value to the machine learning-based analysis model (S2). At this time, the machine learning-based analysis model is learned through a training image in which the type of object is known in advance, and has a function of detecting an object in an unknown image using parameters based on the learning result, for example. It may be to use a convolutional neural network (CNN), but is not limited thereto.

At this time, the image segmentation unit 32 specifies a boundary image in which the type and/or number of objects detected from the image by the machine learning unit 33 changes significantly beyond a preset level (S3), and uses the boundary image to Image information may be divided into unit image data so that before and after the boundary image becomes separate unit image data (S4).

3 is a conceptual diagram of a process of dividing unit image data by applying object detection by machine learning in a block chain-based image forgery prevention method according to an embodiment.

Referring to FIG. 3 , the image information 100 in the form of a moving picture may be composed of a series of images 111 to 115 and 121-125 consecutively in time series with each other. In this case, the block chain-based image forgery prevention system can determine the type of object included in each image, the area in the image where the object is located, the number of objects included in the image, and the like through machine learning-based analysis. For example, the image 115 represents an image including an object 1150 corresponding to an adult male, and the image 121 includes an object 1210 corresponding to an adult female and an object 1211 corresponding to a child. displayed image.

At this time, the image 115 and the image 121 are images that are temporally continuous with each other, but as shown in the drawing, the object 1150 included in the preceding image 115 and the object included in the following image 121 are The types and numbers of (1210, 1211) have changed. In the actual captured image, there will be many cases where the detected object changes more gradually, but for convenience of explanation, it is easy for those skilled in the art that the extreme form in which the type and number of objects vary in one image interval is exemplified. will be understood

In the example of FIG. 3 , since the image 115 and/or the image 121 in which the type and number of objects greatly change correspond to the boundary image, the

unit image data

11 and 12 may be defined using this as a division reference point. For example, the section from the start of the image information 100 to the image 115 becomes the first unit image data 11, and the section starting from the next image 121 until another boundary image is detected is the second. The image information 100 may be divided into a plurality of unit image data to become two unit image data 12 .

In the example shown in FIG. 3 , the types of objects (adult male, adult female, child) and the number of objects included in the image of each unit image data are all different as an example, but in some embodiments, the number of objects is the same. However, it is also possible to specify as a boundary image images in which the types of objects are different or the number of objects is different even though the types of objects are the same.

Referring back to FIGS. 1 and 2 , as described above, for each unit image data partitioned based on the number of frames or time length, or based on object detection, the block chain-based image forgery prevention system 3 The hash extraction unit 34 may calculate a hash value by applying a hash function (S5). As a hash function for calculating the hash value, for example, Secure Hash Algorithm (SHA) may be used, but is not limited thereto.

Next, the storage unit 36 of the block chain-based image forgery prevention system 3 may record the hash value of the unit image data in a data block on the block chain network 2 (S8). At this time, the storage unit 36 of the block chain-based image forgery prevention system 3 generates transaction information corresponding to each unit image data (S7), and transmits the transaction information along with the hash value of the unit image data to the blockchain network. (2) It is also possible to write to the data block above (S8).

Referring to FIG. 4 , the block chain may consist of a series of data blocks 301-303 consecutive to each other, and each data block 301-303 includes counter information indicating the serial number of the corresponding data block in the distributed ledger. (3013). In addition,

hash values

3011, 3021, 3031 of the unit image data assigned thereto are recorded in each data block 301-303, and also hash

values

3012, 3022, 3032 of the preceding data block on the block chain. This is recorded. For example, each of the data blocks 301-303 stores hash

values

3011, 3021, and 3031 of unit image data that are continuous with each other, and a preceding data block (for example, 302) For example, a hash value 3022 of 301 is stored.

Since the hash value is changed to a completely different value when any part of the data is changed due to the nature of the hash function, when the image data hash value 3011 included in the preceding block 301 is changed due to the modulation of image information, the corresponding data The hash value of block 301 is changed to a completely different value. In addition, since the hash value 3022 of the previous block of the data block 302 that follows the data block 301 is changed as a result, the hash value of the data block 302 that follows is also changed and the block chain in this way Hash values of all blocks after the block corresponding to the modulated image data are changed. Therefore, by storing the hash value of the image data in the block chain, it is possible to detect that the image is manipulated or forged.

In each data block 301-303,

transaction information

3014, 3024, 3034 related to the movement of a token whose issuance and deposit/withdrawal are managed by the distributed ledger on the block chain may be further recorded. In an embodiment, the

transaction information

3014, 3024, and 3034 may be defined by reflecting the characteristics of unit image data corresponding to the image

data hash values

3011, 3021, and 3031 of each data block 301-303. . For example, the movement direction related to the movement of the token defined by the

transaction information

3014 , 3024 , 3034 may be determined according to the characteristics of the image data, the sender and the receiver, the movement quantity, etc. may be determined, which will be described later in detail.

Referring back to FIGS. 1 and 2 , in one embodiment, the storage unit 36 of the block chain-based image forgery prevention system 3 stores the hash value of the unit image data on the block chain, and the same image information may be received from different photographing apparatuses 1 and cross-verifying blocks in which hash values of unit image data generated from image information received from each photographing apparatus 1 are recorded (S6). To this end, each of the photographing devices 1 may be configured to be able to communicate with each other through a wired and/or wireless network, and the block chain-based image forgery prevention system 3 may include any one or both of these photographing devices ( 1), or may be implemented as separate devices capable of communicating with all of them.

An image photographed by any one photographing device 1 is input to the block chain-based image forgery prevention system 3 provided in or linked to the corresponding photographing apparatus 1, while another photographing is performed through communication through a network can be transferred to the device 1 . In another photographing device 1, processing of image information is made through the block chain-based image forgery prevention system 3 provided or linked thereto. In this case, hash value generation of unit image data and generation of data blocks are respectively performed on the basis of the image information of both photographing devices 1 . When there is no manipulation or forgery of image information, a hash of a series of generated data blocks The values should be the same on both devices. Therefore, when the hash values of the data blocks are cross-verified, if there is a difference in the hash values of the data blocks generated by both devices, it can be known that manipulation of the data exists.

In addition, in one embodiment, transaction information is recorded in a data block together with a hash value of unit image data, but transaction information is defined by reflecting the characteristics of unit image data, so that monitoring, management, or monitoring of image information is performed through transaction information. can make it possible In this case, the characteristic of the unit image data refers to the type and/or number of objects detected based on machine learning from images constituting the unit image data, and information that can specify a user related to the image information is received along with the image information. If it is, it may refer to user information corresponding to the unit image data.

For example, the transaction generating unit 362 of the block chain-based image forgery prevention system 3 defines a transaction corresponding to each unit of image data of image information as a movement of a token, but at this time, It may be determined based on the presence, type, and/or number of objects identified from images included in the unit image data. For example, a transaction is defined that moves a token from a preset sending address to a preset receiving address corresponding to each data block, but when the number of objects detected from the image of the unit image data is 1, a token shift is generated by 1 and, when the number of objects detected from the image of the unit image data is 2, token movement by 2 may be generated. If the token movement quantity is defined according to the object detection result as described above, the inflection point at which the number of objects in the image changes can be easily detected through the change amount per transaction of the token quantity received at the receiving address.

As another example, the transaction generating unit 362 defines a transaction for moving a token from a sending address to a receiving address corresponding to each unit of image data of the image information, but whether an object is detected from the image of the unit image data, the object to be detected It may be configured to change the movement direction of the token according to the type of the token and/or the number of detected objects. For example, if an object is detected from the image of unit image data, the token is moved from the sending address to the receiving address, but when the object is not detected from the image, the transaction is executed to move the token by swapping the sending address and the receiving address. can be defined Alternatively, when a first type of object (eg, male) is detected from the image of the unit image data, the token is moved from the sending address to the receiving address, and when a different second type of object (eg, female) is detected Transactions can be defined to move tokens by swapping the sending and receiving addresses. If the token movement quantity is defined according to the object detection result as described above, an inflection point at which the type or presence of an object in the image changes can be easily detected through the increase or decrease of the token quantity received at the receiving address.

As another example, the user management unit 37 receives the user information of a user who has photographed the corresponding image or manages the corresponding image together with the reception of the image information through the input unit 31, and the transaction generating unit 362 is each unit image By reflecting user information corresponding to the data, it is possible to define a transaction to be recorded in the same data block as the hash value of the corresponding unit image data. For example, the user management unit 37 stores the e-wallet addresses of each user for a plurality of preset users, and the transaction generation unit 362 records the hash value of each unit image data in the data block. With the user's e-wallet address corresponding to the unit image data as the sending address or the receiving address, the transaction defining the movement of the token can be recorded together. In this case, personal information such as an administrator or photographer of an image can be easily identified by only moving the token through the wallet address of the sender who transmits the token through the transaction or the sender who transmits the corresponding token.

5 is a conceptual diagram for explaining video monitoring through a blockchain-based token transaction in a block-chain-based image forgery prevention method according to an embodiment. is shown according to At this time, the number of tokens was changed by the transaction recorded in the data block along with the hash value of the image data.

As shown in FIG. 5 , the token quantity constantly increased over time in the first section 501 , but the slope at which the tokens increased in the next section 502 changed, and thereafter, in the next section 503 , tokens again This increasing slope decreased. This means that if a first amount of tokens per hour is deposited in the

sections

501 and 503 , in the section 502 , a second amount of tokens per hour greater than the first amount are deposited. When the amount of token movement through a transaction is defined according to the type or number of objects in the image of the image data, the amount of token movement per hour between

sections

501 and 502 and between

sections

502 and 503 Through this change, it can be seen that a change has occurred in the object detected at the corresponding inflection point.

In addition, in the section 501-503, the token quantity of the receiving address increased over time, which means that tokens were continuously deposited from other sending addresses to the corresponding receiving address. On the other hand, in the subsequent section 504 , the token quantity of the receiving address decreases over time, which means that the receiving address is no longer a receiver but a sender, and token withdrawal is made from the corresponding address. Therefore, when the movement direction of the token through the transaction is defined according to the type or number of objects in the image of the image data, the object detected at the corresponding inflection point through the change in the form of the transaction between the section 503 and the section 504 It can be seen that changes have occurred in

Referring back to FIGS. 1 and 2 , in an embodiment, the storage unit 36 may further store one or more unit image data constituting image information or the image information itself ( S9 ). Although the process (S9) of storing image information in FIG. 2 is illustrated as being performed after the process of storing a hash value for preventing forgery and falsification of image information, this is an example for convenience of explanation, and a hash value for preventing forgery and falsification of image information Image information may be stored in the storage prior to or in parallel with the storage of the hash value.

According to the block chain-based image forgery prevention method according to the embodiments described above, image data such as a moving image file is divided into predetermined unit image data, and a hash value of each unit image data is stored in a data block on a block chain network by time. By storing in a chain form, if there is forgery or manipulation of the image data, it is possible to recognize that the image has been manipulated through the fact that the hash value of the image data is different from the hash value stored on the block chain. Through this, forgery of image data can be prevented, and reliability can be increased when the image is used as evidence for illegal activities such as crimes and incidents.

The operation by the block chain-based image forgery prevention method according to the embodiments described above may be implemented at least partially as a computer program and recorded in a computer-readable recording medium. A program for implementing the operation by the block chain-based image forgery prevention method according to the embodiments is recorded and the computer-readable recording medium includes all kinds of recording devices in which computer-readable data is stored. . Examples of the computer-readable recording medium include ROM, RAM, CD-ROM, magnetic tape, floppy disk, and optical data storage device. In addition, the computer-readable recording medium may be distributed in network-connected computer systems, and the computer-readable code may be stored and executed in a distributed manner. In addition, functional programs, codes, and code segments for implementing the present embodiment may be easily understood by those skilled in the art to which the present embodiment belongs.

Although the present invention as described above has been described with reference to the embodiments shown in the drawings, it will be understood that these are merely exemplary, and that various modifications and variations of the embodiments are possible therefrom by those of ordinary skill in the art. However, such modifications should be considered to be within the technical protection scope of the present invention. Accordingly, the true technical protection scope of the present invention should be defined by the technical spirit of the appended claims.

Claims

an input unit configured to receive image information from a photographing device;

an image dividing unit configured to divide the image information into a plurality of unit image data;

a hash extraction unit configured to calculate a hash value of each of the plurality of unit image data; and

A hash value of each of the plurality of unit image data is recorded in data blocks on a blockchain network, wherein each of the data blocks includes a hash value of a preceding data block and a hash value of the unit image data, Block chain-based image forgery prevention including a storage unit configured to record a hash value of unit image data preceding each unit image data corresponding to each data block among unit image data in a data block preceding each data block system.
According to claim 1,

Further comprising a machine learning unit configured to detect an object corresponding to the images by using the images included in the image information as input values for a machine learning-based analysis model,

The image dividing unit specifies an image in which at least one of a type of an object detected by the machine learning unit and the number of detected objects is changed, and divides before and after the specified image in the image information into different unit image data A blockchain-based video forgery prevention system further configured to
3. The method of claim 2,

The storage unit is further configured to record transaction information for a token on the block chain network defined corresponding to the unit image data in data blocks on the block chain network together with the unit image data. system.
4. The method of claim 3,

The storage unit, based on the type or number of objects detected by the machine learning unit with respect to the image included in the unit image data, the entry/exit details of the token defined by the transaction information or the token defined by the transaction information A blockchain-based video forgery prevention system further configured to determine one or more of the quantities.
4. The method of claim 3,

Including a wallet address information for each user registered in advance, further comprising a user management unit configured to receive user information corresponding to at least a part of the image information,

The storage unit is further configured to determine a wallet address for the unit image data corresponding to each data block based on the user information, and to define the transaction information using the determined wallet address as a sender or a recipient. video forgery prevention system.
According to claim 1,

The storage unit, a block chain-based image forgery prevention system further configured to store the image information in the image storage system or an external storage.
Block chain-based image forgery prevention system receiving image information from a photographing device;

dividing, by the block chain-based image forgery prevention system, the image information into a plurality of unit image data;

calculating, by the block chain-based image forgery prevention system, a hash value of each of the plurality of unit image data; and

Comprising the step of the block-chain-based image forgery prevention system recording the hash value of each of the plurality of unit image data in data blocks on a block chain network,

Each of the data blocks includes a hash value of the preceding data block and a hash value of the unit image data,

In the step of writing to the data blocks on the block chain network, the block chain-based image forgery prevention system is unit image data preceding the unit image data corresponding to each data block among the plurality of unit image data. A block chain-based image forgery prevention method comprising the step of recording the hash value of the data block preceding each data block.
8. The method of claim 7,

Before the step of dividing the plurality of unit image data, the block chain-based image forgery prevention system uses the images included in the image information as input values for a machine learning-based analysis model to correspond to the images. Further comprising the step of detecting the object,

The step of dividing into the plurality of unit image data comprises:

specifying, by the block chain-based image forgery prevention system, an image in which at least one of a type and number of detected objects is changed by the machine learning-based analysis model; and

The block-chain-based image forgery prevention method comprising the step of, by the block-chain-based image forgery prevention system, dividing before and after the specified image in the image information into different unit image data.
9. The method of claim 8,

Recording, by the blockchain-based image forgery prevention system, transaction information for a token on the blockchain network defined corresponding to the unit image data in data blocks on the blockchain network together with the unit image data Block chain-based video forgery prevention method comprising more.
10. The method of claim 9,

The step of recording the transaction information in the data blocks on the block chain network is that the block chain-based image forgery prevention system detects the image included in the unit image data by the machine learning-based analysis model. Based on the type or number of objects, the block chain-based image forgery prevention method comprising the step of determining at least one of the token entry/exit details defined by the transaction information or the quantity of tokens defined by the transaction information.
10. The method of claim 9,

storing wallet address information for each user registered in advance in the block chain-based image forgery prevention system; and

The block chain-based image forgery prevention system further comprises the step of receiving user information corresponding to at least a part of the image information,

The step of writing the transaction information to data blocks on the blockchain network comprises:

determining, by the block chain-based image forgery prevention system, a wallet address for the unit image data corresponding to each data block based on the user information; and

and defining, by the block-chain-based video forgery prevention system, the transaction information using the determined wallet address as a sender or a recipient.
8. The method of claim 7,

The block-chain-based image forgery prevention method further comprising the step of, by the block-chain-based image forgery prevention system, storing the image information in the image storage system or an external storage.
A computer program stored in a computer-readable recording medium to execute the block chain-based image forgery prevention method according to any one of claims 7 to 12 in combination with hardware.