KR102178313B1 - Image analysis apparatus and system comprising the same - Google Patents

Abstract

The present invention relates to an image analysis system. The image analysis system comprises: a smart contract device receiving an analysis image, creating a text information form for analysis on an analysis image in accordance with a smart contract with contract conditions, generating text information for analysis for the analysis image, and storing the text information in the block-chain main net; and a data analysis device extracting at least one text information for analysis suitable for an applied search condition through the block-chain main net, calculating a summation result for each contract condition by summing the extracted at least one text information for analysis with the result value for the same contract condition, and determining the truth of each contract condition using the summation result to generate an analysis result of the extracted at least one text information for analysis.

Description

An image analysis device and an image analysis system including the same TECHNICAL FIELD [IMAGE ANALYSIS APPARATUS AND SYSTEM COMPRISING THE SAME}

The present invention relates to an image analysis device and an image analysis system including the same, and more particularly, to an image analysis device that reduces the amount of processed data of an acquired image and manages an acquired image in a block chain method, and an image analysis system including the same. It is about.

In order to prevent crime and reduce the incidence of accidents, the installation of image acquisition devices such as CCTV (closed circuit television) is increasing day by day.

In general, CCTV is installed in a desired place in a public institution such as a police station, and was mainly used to detect the cause of a criminal or traffic accident.

However, these days, individuals or private companies are installing a lot of CCTV around their homes or businesses in order to protect themselves or prevent theft.

The CCTV installed in this way captures and stores the surrounding area in real time.

These days, due to the development of imaging technology or analysis technology, the quality of captured images is improving, but many of the CCTVs currently installed still have low resolution, causing a lot of difficulty in image analysis. In many cases, the desired information cannot be confirmed through the CCTV video.

Therefore, in order to smoothly perform the video providing service using CCTV video, it is often necessary to obtain the video obtained from CCTV installed by individuals and private companies as well as public institutions. In this case, the video must be secured through the CCTV owner.

However, in the case of multiple CCTV images acquired through such a process, integrity in which data storage in a normal state is secured without manipulation or damage must be guaranteed.

However, in the case of current CCTV images, it is not possible to verify damage or manipulation, and there are many difficulties in processing and storing images captured in real time.

In addition, the CCTV images that have been captured and are currently being stored can be manipulated through hacking or the like, and thus the accuracy of the acquired CCTV images may not be assured.

Republic of Korea Patent Publication No. 10-1513215 (Notification date: April 17, 2015, title of invention: Object behavior pattern CCTV image analysis server)

The problem to be solved by the present invention is to improve the reliability of the image by securing the integrity of the acquired image.

Another problem to be solved by the present invention is to shorten the image processing time by reducing the data processing amount of the image and to secure the ease of storage.

An image analysis system according to one aspect of the present invention for solving the above problem receives an analysis image and creates a text information form for analysis for the analysis image according to a smart contract with contract conditions to analyze the analysis image. A smart contract device that creates text information for and stores it in the blockchain mainnet, and at least one analysis text information that meets the authorized search conditions is extracted through the blockchain mainnet, and the extracted at least one analysis Calculate the summation result for each contract condition by adding text information to the result value for the same contract condition, and use the summation result to determine the truth about each contract condition and extract at least one text information for analysis. It includes a data analysis device for generating an analysis result.

The video analysis device determines the number of votes for the extracted text information for analysis, applies the number of votes to the result value for each contract condition of the extracted text information for analysis, and then determines the result value to which the number of votes is applied. Can be used to sum the results for the same contract terms.

The number of votes may be determined using at least one of owner information about the owner of each extracted text information for analysis, specifications and integrity information of an image acquisition device owned by the owner.

The integrity information may include whether an image is damaged in an image acquisition field providing text information for each analysis and the number of image damage.

The search condition may be at least one of a date, a time, a location range, a feature of a subject to be searched, owner information, a specification of an image acquisition device, and weather information.

A data analysis device according to another aspect of the present invention includes a user input unit for entering a search condition, a storage unit in which the number of votes for each owner is stored, a communication unit that accesses the blockchain mainnet and receives at least one text information for analysis, And at least one analysis text information that is connected to the user input unit, the storage unit, and the communication unit and meets a search condition input through the user input unit is extracted from the blockchain mainnet through the communication unit, and the extracted The summation result for each contract condition is calculated by summing the result value for the same contract condition of at least one of the text information for analysis, and at least one extracted by determining the truth about each contract condition using the summation result. And a control unit generating an analysis result for the text information for analysis.

The control unit determines the number of votes for the extracted text information for analysis before being added to the result value for the same contract condition, and the result value for each contract condition of the extracted text information for analysis. After applying to, the result values for the same contract terms can be summed using the result value applied with the number of votes.

The number of votes may be determined using at least one of owner information about the owner of each extracted text information for analysis, specifications and integrity information of an image acquisition device owned by the owner.

The data analysis device according to the feature may further include an information output unit connected to the control unit and outputting the analysis result or a ratio of the summation result for each contract condition.

According to this feature, the result of each contract condition for analysis text information is voted, that is, summed to determine the truth of each contract condition, and accordingly, it is easy to determine whether the text information for analysis is damaged.

For this reason, accurate information analysis is performed using accurate text information for analysis that is not damaged.

In addition, instead of storing the analyzed image provided from each image capturing device, information on the analyzed image is converted into a text format and stored, so the amount of stored and processed data is greatly reduced.

Since the image itself is stored or text information for analysis converted from the corresponding image into text information is stored, privacy of individuals exposed to the image is protected.

Moreover, since text information for analysis is stored using a block chain, the integrity of the text information for analysis is greatly increased.

1 is a schematic block diagram of an imaging system according to an embodiment of the present invention.
2 is a flowchart illustrating an operation of an image analysis apparatus according to an embodiment of the present invention.

Hereinafter, embodiments of the present invention will be described in detail with reference to the accompanying drawings. In describing the present invention, if it is determined that adding a detailed description of a technology or configuration already known in the relevant field may make the subject matter of the present invention unclear, some of these will be omitted from the detailed description. In addition, terms used in the present specification are terms used to appropriately express embodiments of the present invention, and these may vary according to related people or customs in the field. Accordingly, definitions of these terms should be made based on the contents throughout the present specification.

The terminology used herein is for reference only to specific embodiments and is not intended to limit the invention. Singular forms as used herein also include plural forms unless the phrases clearly indicate the opposite. As used in the specification, the meaning of'comprising' specifies a specific characteristic, region, integer, step, action, element and/or component, and other specific characteristic, region, integer, step, action, element, component and/or group It does not exclude the existence or addition of

Hereinafter, an image analysis apparatus and an image analysis system including the same according to an embodiment of the present invention will be described with reference to the accompanying drawings.

First, an image system including an image analysis system according to an embodiment of the present invention will be described with reference to FIG. 1.

As shown in FIG. 1, the video system of this example includes a plurality of image acquisition devices 10, a smart contract device 20 communicating with each image acquisition device 10, and a blockchain main device communicating with the smart contract device 20. A data analysis device 40 that communicates with the net 30 and the blockchain main net 30 is provided.

Each image acquisition device 10 is a device that acquires an image around a location where it is installed in real time using a camera or the like, and may be, for example, a CCTV.

As shown in FIG. 1, the image acquisition device 10 includes an image acquisition unit 11, a storage unit 12 connected to the image acquisition unit 11, and an image analysis connected to the storage unit 12. A communication unit 14 connected to the unit 13 and the image analysis unit 13 is provided.

The image acquisition unit 11 is a part that photographs a corresponding region and generates an image of the photographed region, and may be a digital camera.

The storage unit 12 is a storage medium that stores an image applied from the image acquisition unit 11 in a data format. In this case, the storage unit 12 may store an image in units of frames.

For this reason, the image analysis unit 13 may perform image analysis using the image stored in the storage unit 12.

The image analysis unit 13 connected to the storage unit 12 analyzes the image stored in the storage unit 12 and outputs an analyzed image (ie, an analysis image).

At this time, the image analysis unit 13 performs an image analysis operation according to analysis information applied from the outside.

Analysis information is information for extracting only an image of a desired part from a captured image, and may include a behavior pattern of a desired subject, an image of a part to be extracted (e.g., a human face or a car), color, analysis date and time, etc. have.

Therefore, a user who wants to analyze an image captured by the image acquisition unit 11 may use a keyboard or mouse to analyze (ie, extract) at least one piece of analysis information for analyzing (ie, extracting) only the image he/she wants from the captured image. ) Is input to the image analysis unit 13 using a user input unit (not shown).

Accordingly, the image analysis unit 13 extracts an image corresponding to the input analysis information from the image stored in the storage unit 13 and transmits it to the smart contract unit 20 through the communication unit 14.

As described above, the communication unit 14 is for wired/wireless communication with an external device (eg, smart contract device 20), and may include an Internet communication module for Internet communication.

In this example, the image analysis unit 13 uses the communication unit 14 to generate a smart analysis image in real time or at a predetermined period, or whenever a predetermined event (eg, a specific face or a car of a specific color) occurs through analysis information. It can be transmitted to the contract unit 20.

In this example, the owner of each image acquisition device 10 may be registered as a member of a website operated by the smart contract device 20 through online or the like using his terminal (ie, the owner terminal).

In this example, the smart contract device 20 may be a server such as a web server, and may transmit and receive cryptocurrency such as Ethereum through the owner terminal.

Such a smart contract device 20 creates a smart contract (i.e., a text information form for analysis) for the analysis image transmitted from each image acquisition device 10 through a smart contract based on a block chain. The text information for analysis is vivid, and the hash valve for the analysis image is extracted, and the text information for analysis and the hash value are transmitted to the blockchain mainnet 30.

A smart contract is a concept in which code is executed when a specific programmed condition (i.e., contract condition) is reached based on Distributed Ledger Technology (DLT) (e.g., blockchain).

The smart contract for such a smart contract is an application program that specifies the type of information collected by automatically indicating whether the smart condition is satisfied in this example, and may be implemented as Ethereum as an example.

In this example, an example of the text information form for analysis is shown in [Table 1] below, and an example of the text information for analysis generated according to the text information form for analysis in [Table 1] is shown in [Table 2]. .

In [Table 1] and [Table 2], the type of image acquisition device and at least one contract condition are provided, and in this case, the contract condition is whether a specific car (Car A) exists, whether a specific person (Person A) exists, fights It may be whether it is (Fight), whether it is dumping, and whether or not there is a crash. Therefore, as an example, the result value for each contract condition may be as shown in [Table 2].

Car A Person A Fight Dumping Crash

Car A Person A Fight Dumping Crash CCTV A Yes No No No Yes

When a specific analysis image is analyzed using a smart contract and text information for analysis having a result value for each contract condition is obtained as shown in [Table 2], the identification information of the image acquisition device from which the analysis image was obtained is CCTV A. , It can be seen that'car A'exists in the analysis video (Yes), and'person A'does not exist. In addition, it can be seen that'Fight','Dumping', and'Crash' did not occur in the analysis image (No).

For this operation, the smart contract device 20 includes a communication unit 21, a storage unit 22, and a control unit 23 connected to the communication unit 21 and the storage unit 22.

The communication unit 21 is also for communication with an external device and may also include an Internet communication module.

Accordingly, the smart contract device 20 receives an analysis image from each image acquisition device 10 through the communication unit 21.

The storage unit 22 stores data required for the operation of the smart contract device 20 or data generated during the operation, and also stores an analysis image transmitted from each image acquisition device 10. In addition, owner information can be stored as needed.

In this case, the owner information may include at least one of a name, an age, a contact number, identification information of an image acquisition device owned, and an owner address (eg, an Ethereum address) for payment of cryptocurrency.

The image acquisition device information may include specification and integrity information of the image acquisition device 10. In this case, the specification of the image acquisition device 10 may include at least one of an installation date, a model name and a manufacturer, an installation location, a photographing range, a resolution, and whether or not to recognize voice. In addition, the integrity information may include whether or not the image is damaged and the number of times the image is damaged by the image acquisition device 10, that is, the image acquisition device 10 providing an analysis image for the text information for analysis.

The control unit 23 generates text information for analysis for analyzing the analyzed image transmitted from each image acquisition device 10 and a hash value for the analyzed image according to the programmed smart contract.

Then, the control unit 23 stores text information for analysis and a hash value for the analysis image in the blockchain mainnet 40.

The blockchain mainnet 40 can have its own platform, and can be the main net of Ethereum.

Accordingly, the text information for analysis and the hash value stored in the blockchain mainnet 30 are stored in the form of a block chain, so that the integrity of the text information for analysis and the hash value is guaranteed.

In this way, instead of storing the analysis image itself provided from each image capturing device 10 in the blockchain mainnet 40, information on the analysis image is converted and stored in a text format, so that the blockchain mainnet 40 The amount of data stored in) is significantly reduced.

In addition, when the analysis image itself is stored, even private information is exposed, but as the analysis image is stored in a text format, concerns about private life exposure decrease. Thus, the anxiety and discomfort felt by individuals captured in the analysis image disappear.

The data analysis device 40 analyzes text information for analysis stored in the blockchain mainnet 30 to obtain desired information, and also checks whether the text information for analysis is damaged using the hash value.

The data analysis device 40 includes a communication unit 41, a user input unit 42, a control unit 43 connected to the communication unit 41 and the user input unit 42, and a storage unit 44 connected to the control unit 43. ) And an information output unit 45 connected to the control unit 43.

The communication unit 41 is for communication with the block chain mainnet 40, and includes an Internet communication module.

The user input unit 42 is for inputting a command or a search condition for the operation of the data analysis device 40, and generates a corresponding signal according to the user's input operation and outputs it to the control unit 43. Accordingly, the control unit 43 determines an operation of the user input unit 42 using a signal applied from the user input unit 42 and executes a corresponding operation or command.

The control unit 43 is a control unit that controls the overall operation of the data analysis device 40, and controls operations such as extraction of desired information by searching text information for analysis that meets the search conditions input through the user input unit 42. do.

The storage unit 44 stores data necessary for the operation of the data analysis device 40 or data generated during the operation, and may store an analysis result that is a result of the analysis operation of the control unit 43, for example. .

The information output unit 45 operates under the control of the control unit 43 to output data output from the control unit 43. The information output unit 45 may be a display device such as a liquid crystal display device or an organic light-emitting display device, and in some cases may include an audio output unit.

Accordingly, a user who uses the data analysis device 40 checks the analysis result by using the data output through the information output unit 45.

In this example, the smart contract device 20 and the data analysis device 40 are composed of separate components from each other, but are not limited thereto and consist of one component.

The operation of the data analysis device 40 having such a structure will be described with reference to FIG. 2.

When power required for the operation of the data analysis device 40 is supplied, the control unit 43 of the data analysis device 40 starts the operation (S10).

Accordingly, the control unit 43 reads the signal applied from the user input unit 42 and determines a search condition input by the user (S11).

The search condition is at least one of date, time, location range, characteristics of the subject to be searched (for example, the type of subject such as a car or a person or impression, etc.), owner information, specifications of the image acquisition device 10, and weather information. Can be

In this way, when a search condition is input through the user input unit 42, the control unit 44 connects to the blockchain mainnet 30, and through the blockchain mainnet, at least one The text information for analysis of is extracted (S12).

Accordingly, the data processing amount of the data analysis apparatus 40 is greatly reduced, so that the data analysis time is greatly shortened, and the accuracy of the data analysis result, that is, the accuracy of the search is also greatly improved.

In this way, when the text information for analysis matching the desired search condition is extracted (S12), the control unit 43 reads the number of votes (ie, weight) for the extracted text information for analysis from the storage unit 44 (S13). ).

The number of votes for the text information for analysis is at least among the owner information of the image acquisition device 10 that provides an analysis image for each analysis text information, and the specifications and integrity information of the image acquisition device 10 owned by the corresponding owner. It can be determined using one.

For example, as the number of image acquisition devices owned by the corresponding owner increases, the number of votes may increase, and as the installation period of the image acquisition device increases, the number of votes may increase.

In addition, the higher the specification of the image acquisition device 10, such as whether the resolution of the image acquisition device, manufacturer or voice recognition, the higher the accuracy of the image captured by the image acquisition device 10, so the specification of the image acquisition device 10 As this increases, the number of votes for textual information for analysis may also increase.

In addition, the number of votes may be selected using the integrity information of the image acquisition device 10 corresponding to the text information for analysis. In this case, the number of votes increases as text information for analysis with high integrity increases.

In this way, when the number of votes for each extracted analysis text information is determined, the control unit 43 applies the number of votes to each row of the corresponding analysis text information, that is, the result value of each contract condition, The final result is determined (S14).

For example, it is assumed that the number of extracted text information for analysis is a total of 4 as shown in [Table 3] below, and the result value of each contract condition is as follows.

Car A Person A Fight Dumping Crash CCTV A Yes No No No Yes CCTV E No Yes No Yes No CCTV F No Yes No Yes No CCTV G No Yes No No No

In this text information for analysis, if the number of votes for CCTV A and G is 1, the number of votes for CCTV F is 1.2, and the number of votes for CCTV E is 1.3, the result of applying the corresponding number of votes to the result is [ Table 4].

Car A Person A Fight Dumping Crash CCTV A Yes 1No 1No 1No 1Yes CCTV E 1.3No 1.3Yes 1.3No 1.3Yes 1.3No CCTV F 1.2No 1.2Yes 1.2No 1.2Yes 1.2No CCTV G 1No 1Yes 1No 1No 1No

In this way, when the number of votes of the value corresponding to each result value of the analysis text information is applied (S14), the control unit 43 is located in the same column and sums the result values having the same contract conditions for each column. For example, in [Table 4], the summation results for the result values of each column are shown in [Table 5].

Car A Person A Fight Dumping Crash Yes count One 3.5 0 2 One No count 3.5 One 4.5 2.5 3.5

Therefore, the control unit 43 generates an analysis result for the text information for analysis extracted by using the summation result for each contract condition, and determines whether the text information for analysis is damaged using the analysis result, and the desired accurate information Is extracted (S16).

For example, if the summation result as shown in [Table 5] comes out, the control unit 42 for each contract condition does not exist Car A (No), Person A exists (Yes), fights (Fight), garbage It is determined that dumping and crashing did not occur (No) as'true, that is,'truth', and the result determined as'truth' is generated as an analysis result.

Therefore, the control unit 42 uses the analysis result as a result of'True', that is, Car A does not exist (No), Person A exists (Yes), fights, dumping, and collision. Analysis information for text having a contract condition in which at least one of no crash has occurred (No) has a different result is determined as analysis information for text that is damaged or damaged.

In this way, when the analysis result for the extracted text analysis information is displayed, the control unit 43 outputs the analysis result using the information output unit 45 (S16).

At this time, the control unit 43 is the result of the analyzed analysis generated, for example, Car A does not exist (No), Person A does not exist (Yes), fights (Fight), garbage dumping (Dumping) and collision It is possible to output that (Crash) did not occur (No), but in contrast, the ratio of the sum result for each contract condition, for example, the first result [Yes] and the first result for each contract condition. 2 It can be output as a ratio of result [No] (eg, 1:3.5).

In addition, the control unit 43 immediately outputs the determined analysis result when the difference between the summation results having contradictory results under the same contract condition is greater than or equal to the set value, and otherwise, the control unit 43 The sum result can be output as a percentage.

Accordingly, using the result output from the information output unit 45, the user can immediately take a necessary operation or action.

For example, when the analysis result is immediately output to the information output unit 45 and it is confirmed that a collision has occurred in a specific area, police officers and medical staff can be dispatched to the corresponding area immediately, and through the audio output unit A dispatch order or the like may be output.

However, when the result of the summation of information is output to the information output unit 45 at a ratio, the person concerned visually checks the image of the corresponding area directly to check whether a collision has occurred, and decides whether to move to the corresponding area.

Due to this, economic and time loss caused by unnecessary dispatch due to wrong judgment can be prevented.

In addition, the data analysis apparatus 40 of the present example may calculate a contribution according to the achievement of the purpose of the person concerned, such as arresting a criminal among the extracted text information for analysis, and pay other compensation to the corresponding owner. At this time, it is already stored in the storage unit 44 as a compensation standard, and the compensation may be made of a cryptocurrency such as Ethereum.

Therefore, when the compensation target and the compensation amount are determined according to the compensation criteria set in the storage unit 44, the control unit 43 searches for the owner of the target text information for analysis, and uses the searched owner information on the owner. The corresponding amount of cryptocurrency can be paid to the owner's address.

In this example, the smart contract device 20 and the data analysis device 40 constitute the image analysis system of this example, and as shown in FIG. 1, the smart contract device 20 and the data analysis device 40 are separate from each other. It consists of components.

However, unlike this, the smart contract device 20 and the data analysis device 40 may be configured as one module.

In the above, embodiments of the image analysis apparatus and the image analysis system including the same have been described. The present invention is not limited to the above-described embodiments and the accompanying drawings, and various modifications and variations are possible from the perspective of those of ordinary skill in the field to which the present invention pertains. Therefore, the scope of the present invention should be defined by the claims of the present specification as well as those equivalent to the claims.

10: image acquisition device 20: smart contract device
30: blockchain mainnet 40: video analysis device

Claims (9)

A smart contract device that receives an analysis image and creates an analysis text information form for the analysis image according to a smart contract with contract conditions, generates text information for analysis on the analysis image, and stores it in the blockchain mainnet, And
At least one analysis text information that meets the authorized search conditions is extracted through the blockchain mainnet, and the extracted at least one analysis text information is added to the result value for the same contract condition for each contract condition. A data analysis device that calculates a summation result, uses the summation result to determine the truth about each contract condition, and generates an analysis result for the extracted at least one text information for analysis
Data analysis system comprising a. In claim 1,
The data analysis device determines the number of votes for the extracted text information for analysis, applies the number of votes to the result value for each contract condition of the extracted text information for analysis, and then determines the result value to which the number of votes is applied. A data analysis system that sums the results for the same contract terms by using. In paragraph 2,
The number of votes is determined by using at least one of owner information on the owner of each extracted text information for analysis, specifications and integrity information of an image acquisition device owned by the owner. In paragraph 3
The integrity information is a data analysis system including whether an image is damaged by an image acquisition device that provides text information for each analysis and the number of image damage. In claim 1,
The search condition is at least one of a date, a time, a location range, a feature of a subject to be searched, owner information, a specification of an image acquisition device, and weather information. User input section for entering search conditions,
A storage unit that stores the number of votes for each owner,
A communication unit that accesses the blockchain mainnet and receives at least one text information for analysis, and
At least one text information for analysis connected to the user input unit, the storage unit, and the communication unit and meeting a search condition input through the user input unit is extracted from the blockchain mainnet, and the extracted at least The summation result for each contract condition is calculated by summing the result value for the same contract condition of one of the text information for analysis, and at least one extracted by judging the truth about each contract condition using the summation result Control unit that generates analysis results for text information for analysis
Data analysis device comprising a. In paragraph 6,
The control unit determines the number of votes for the extracted text information for analysis before being added to the result value for the same contract condition, and the result value for each contract condition of the extracted text information for analysis. After applying to, a data analysis device that adds up the result values for the same contract terms by using the result value to which the number of votes is applied. In clause 7,
The number of votes is determined using at least one of owner information about the owner of each extracted text information for analysis, specifications and integrity information of an image acquisition device owned by the owner. In paragraph 6,
And an information output unit connected to the control unit and configured to output the analysis result or a ratio of the summation result for each contract condition.

Images