KR20100103221A - Automatic teller machine for preventing illegal finance transaction and method of controlling the same - Google Patents

Abstract

PURPOSE: An automatic teller machine and a method for controlling the same are provided to verify a normal user based on the face image, the face expression, and the body temperature information of a user. CONSTITUTION: A finance transaction request is received through a user interface(210). An image capturing unit(220) captures a user image. An extracting unit(230) extracts a face image from the user image and extracts a face expression from the face image. A body temperature capturing unit(240) captures the temperature information in the face region of the user. A controlling unit(250) verifies a user and determines the response with respect to the finance transaction request.

Description

Financial automation equipment and its control method to prevent illegal financial transactions {AUTOMATIC TELLER MACHINE FOR PREVENTING ILLEGAL FINANCE TRANSACTION AND METHOD OF CONTROLLING THE SAME}

One embodiment of the present invention relates to a financial automation device and a control method thereof, and more particularly, to a financial automation device and a control method thereof that can prevent illegal financial transactions by a user without a right of use.

With respect to traditional financial transactions that have been carried out by banks in direct contact with the clerks, the emergence of Automatic Teller Machines (ATMs) has revolutionized in many ways. The waiting time of the user is significantly reduced, and even after the work of the financial institution, the user can enjoy the benefit of being able to perform financial transactions whenever they want. Not only users, but also financial institutions, processing time for financial services is reduced, and work efficiency is reduced due to a reduction in manpower and costs.

Financial automation equipment basically provides many financial tasks performed by financial institutions such as banks and village safes. The user can perform a desired financial transaction in a short time by simply operating a financial automation device installed in the bank, without going through a bank clerk, for account transfer, balance inquiry, deposit withdrawal, and bankbook arrangement. These advantages of financial automation devices have led to a rapid increase in the number of customers, installations, and installation sites of financial automation devices, making it easy to find them today.

In such a situation in which financial automation devices are rapidly increasing, financial crimes are being used to withdraw cash or checks using another person's bankbook, cash withdrawal card, or credit card. In preparation, financial institutions install unmanned cameras on financial automation devices to record and record the financial transactions of users, and use the recording / recorded recording media to identify criminals' faces in the future. However, this method is effective in the post-discovery of criminals only when the criminal's face is correctly identified, and there is a problem in that financial crime cannot be prevented early.

An object of the present invention is to provide a financial automation device for preventing illegal financial transactions, which can prevent illegal financial transactions by unauthorized users in advance.

Financial automation device for preventing illegal financial transactions according to an embodiment of the present invention is a user interface for receiving a financial transaction request from a user, an image acquisition unit for capturing the user to obtain a user image for the user, the face from the user image An extraction unit for extracting an image and extracting a facial expression from the face image, a body temperature information obtaining unit for obtaining body temperature information of the face region of the user, and based on the face image, the facial expression, and the body temperature information And determining whether the user is a normal user to determine whether to process the financial transaction request.

In this case, the extractor may further extract face material information from the face image, and the controller may determine whether the user is a normal user based on the face material information.

The controller may analyze the behavior pattern of the user from the user image and determine whether the user is a normal user based on the behavior pattern.

The user interface may further include a key input unit configured to receive a key input from the user, and the controller may analyze an input pattern of the key input and determine whether the user is a normal user based on the input pattern. Can be.

In addition, the financial automation device for preventing illegal financial transactions according to another embodiment of the present invention includes a key input unit for receiving a key input from a user, a user interface for receiving a financial transaction request from the user, and photographing the user. An image obtaining unit for obtaining a user image of a user, an extracting unit extracting a face image from the user image, and extracting a facial expression from the face image, and an input pattern of the face image, the facial expression, and the key input And a controller configured to determine whether the financial transaction request is processed by determining whether the user is a normal user based on the determination.

In this case, the extractor may further extract face material information from the face image, and the controller may determine whether the user is a normal user based on the face material information.

In addition, the method for controlling a financial automation device for preventing illegal financial transactions according to an embodiment of the present invention comprises the steps of: receiving a financial transaction request from a user, photographing the user to obtain a user image of the user, and the user image Extracting a face image from the face image, extracting a facial expression from the face image, extracting body temperature information from the face region of the user, and extracting the body image based on the face image, the facial expression, and the body temperature information. Determining whether the user is a normal user, and determining whether to process the financial transaction request according to whether the user is a normal user.

According to the present invention, illegal financial transactions by fraudulent users can be prevented in advance.

Hereinafter, embodiments of the present invention will be described in detail with reference to the accompanying drawings. However, the present invention is not limited or limited by the embodiments. Like reference numerals in the drawings denote like elements.

The term "financial automation device" used in the present invention is not only an ATM (Automated Teller Machine), but also a cash dispenser (CD: cash dispenser) capable of cash withdrawal, various banking, insurance, credit card, etc. It is broadly interpreted as a concept that includes a series of unmanned automated devices such as a kiosk that includes a complex automated device that provides a service and provides various additional services such as utility billing, ticket issuance, and point accumulation. .

1 is a block diagram showing a general structure of a financial automation device according to an embodiment of the present invention.

Reference numeral 110 is an operation terminal for receiving a menu selection and operation for the financial transaction from the user. The operation terminal 110 provides various financial services or additional services to the user through the GUI screen. Reference numeral 120 denotes an entry / exit part of a banknote for counting and storing banknotes, and reference numeral 130 denotes an account book for reading a bankbook and performing a bankbook arranging task. Reference numeral 140 is a card reader that reads a card according to a user's financial transaction card input and performs card related tasks. The card reader 140 may read not only a financial transaction card but also an accumulation card or an ID card for other tasks. Reference numeral 150 is a camera for photographing a user. In addition, the financial automation device may further include a communication unit for accessing the financial automation server, a control unit for controlling the operation of the financial automation device, a memory unit for storing various data, and a printing module for outputting an account book or a statement table. have.

2 is a block diagram illustrating an internal configuration of a financial automation device for preventing illegal financial transactions according to an embodiment of the present invention.

As shown in FIG. 2, the financial automation device 200 for preventing illegal financial transactions according to an embodiment of the present invention includes a user interface 210, an image acquisition unit 220, an extraction unit 230, and a body temperature information acquisition unit. 240 and the controller 250. Here, according to an embodiment of the present invention, the financial automation device 200 for preventing illegal financial transactions may further include a storage unit 260. Hereinafter, the function of each component will be described in detail.

The user interface 210 receives a financial transaction request from a user.

As an example, the user interface 210 may include a key input unit 211 for receiving a key input from a user, a card inlet 212 for inserting a cash or credit card, and a bankbook inlet 213 for injecting a bankbook. Can be. The user interface 210 receives a card or bankbook through the card inlet 212 or the bankbook inlet 213, and receives a financial transaction request by receiving a key input through the key input unit 211. Here, the key input unit 211 may correspond to the operation terminal 110 described with reference to FIG. 1.

The image acquisition unit 220 acquires a user image of the user by photographing the user. The user image acquired by the image acquirer 220 may be at least one.

According to an embodiment of the present disclosure, when a financial transaction request is received from the user through the user interface 210, the image acquirer 220 may start capturing the user to acquire a user image. Here, the image acquirer 220 may include the camera 150 described with reference to FIG. 1.

In addition, the image acquisition unit 220 may acquire a plurality of user images by photographing the user for a predetermined time. Here, the predetermined time is a time from the time when the controller 250 commanded to start photographing to the time when commanded to end photographing, which may be changed by the surrounding environment, conditions, and the like. That is, the image acquisition unit 220 may receive information on the photographing time received from the controller 250, and acquire a plurality of user images by photographing the user during the photographing time.

As an example, when a user inputs a password through the key input unit 211, selects a transaction to be used or inputs an amount to withdraw, or the user enters a card or a bankbook into the card inlet 212 or the bankbook inlet 213. In the case of inputting the image, the image acquisition unit 220 may acquire the user image by photographing the user for a predetermined time. Here, the input of the above-mentioned password, the selection of the transaction, the input of the amount of money, the input of the card, the input of the bankbook and the like are examples of user input through the user interface 210. That is, when any one user input is received from among the plurality of user inputs, the image acquirer 220 may acquire a user image by capturing the user.

In this case, the plurality of user inputs may be sequentially input through the user interface 210. When a plurality of user inputs are sequentially input, the image acquirer 220 may acquire a plurality of user images by photographing the users corresponding to the sequentially input user inputs.

For example, when the user inserts the card into the card slot 212, the image acquisition unit 220 acquires a first user image for the user by photographing the user, after which the user enters the key input unit 211 In the case of inputting a password, the image acquisition unit 220 may acquire a second user image of the user by photographing the user. Similarly, when the user inputs an amount to be withdrawn to the key input unit 211, the image acquisition unit 220 may capture the user to acquire a third user image for the user.

The extractor 230 extracts a face image from the user image acquired by the image acquirer 220, and extracts a facial expression from the extracted face image. When there are a plurality of user images acquired by the image acquirer 220, the extractor 230 extracts a facial expression from all of the plurality of user images.

As mentioned above, the image acquisition unit 220 may acquire a plurality of user images by photographing the user in correspondence with a plurality of sequentially input user inputs. In this case, the image acquisition unit 220 may According to whether the extractor 230 can extract the face image from the user image (for example, the first user image) obtained in response to the priority user input, the user may photograph the user in response to the post rank user input. It may be determined whether to obtain an image (eg, a second user image).

For example, when the first user input is "input of a password", the second user input is "inserting a card", and the first user input and the second user input are sequentially input, The operation is described below.

First, when a password is input (that is, when a first user input is input), the image obtaining unit 220 acquires a first user image by photographing a user.

Thereafter, the extractor 230 extracts a face image from the first user image.

In this case, if the face image is not extracted from the first user image, the image acquisition unit 220 captures the user and captures the second user image when the card is input (that is, when the second user input is input). Can be obtained.

On the contrary, if the face image is extracted from the first user image, the image acquisition unit 220 may not photograph the user even if the card is input (that is, may not acquire the second user image).

The extracted facial image and facial expression are used to determine whether the user who requested the financial transaction is a normal user. Detailed descriptions of the configuration for extracting the facial image from the user image and the configuration for extracting the facial expression from the extracted face image will be described with reference to FIGS. 3 and 4 below.

The body temperature information acquirer 240 acquires body temperature information in the face region of the user.

As an example, the body temperature information acquisition unit 240 may acquire body temperature information in the face region by using a thermal infrared camera or a thermal infrared sensor provided in the financial automation device 200 for preventing illegal financial transactions. The acquired body temperature information is also used to determine whether the user who requested the financial transaction is a normal user, similar to the face image and the facial expression. Detailed description thereof will be described with reference to FIG. 5.

The controller 250 determines whether the requested financial transaction is processed by determining whether the user who requested the financial transaction is a normal user based on the facial image and the facial expression extracted by the extractor. That is, the controller 250 continues the financial transaction when it is determined that the user is a normal user, and stops the financial transaction when it is determined that the user is not a normal user. Normal user means a person who is expected to have a legitimate right to use an account or card associated with the requested financial transaction.

When there are a plurality of user images acquired by the image acquirer 220, the controller 250 determines whether a user who requests a financial transaction is a normal user based on a plurality of face images and a plurality of facial expressions obtained from the plurality of user images. You can judge. In this case, the controller 250 may perform the normal user determination more accurately than when determining whether the normal user is based on one face image and one facial expression.

According to an embodiment of the present invention, the controller 250 determines whether the financial transaction request is processed by determining whether the user who requested the financial transaction is a normal user based on the temperature information extracted by the temperature information extractor 240. Can be.

In addition, according to an embodiment of the present disclosure, the controller 250 may determine whether the user is a normal user based on face material information extracted from the face image of the user. At this time, the face material information is extracted from the extractor 230.

In addition, according to an embodiment of the present invention, the controller 250 may analyze the behavior pattern of the user from the user image, and determine whether the user is a normal user based on the analyzed behavior pattern.

In addition, according to an embodiment of the present invention, the controller 250 analyzes an input pattern of a key input received through the key input unit 211 provided in the user interface 210, and based on the analyzed input pattern It can be determined whether is a normal user.

Hereinafter, the controller 250 determines whether the user who requested the financial transaction is a normal user based on at least one of a face image, a facial expression, body temperature information, facial material information, a user behavior pattern, and an input pattern of a key input. The configuration will be described for each judgment element.

1. Determination of normal user through face image analysis

If the person requesting the financial transaction is a fraudulent user who does not have a legitimate license, the fraudster may use an object to cover the camera or cover his eyes with a hat, sunglasses, or mask to prevent his face from being exposed. Often, financial transactions are covered by covering key components of the face, such as the nose and mouth.

In this case, in order to prevent the financial transaction from proceeding normally, when the main component of the face cannot be detected from the extracted face image, the controller 250 determines that the user who requested the financial transaction is a fraudulent user and requests the requested finance. Stop the transaction.

As one example, when any one of eyes, nose, and mouth are not detected from the face image, the controller 250 may determine that the user who requested the financial transaction is a fraudulent user.

Hereinafter, an example of the present invention for extracting a face image from a user image will be described with reference to FIG. 3.

3 is a diagram illustrating a method of extracting a face image from a user image step by step according to an embodiment of the present invention.

First, edge detection is performed in step S310. That is, in step S310, the edge is detected from the obtained user image to detect the contour of the facial component.

In step S320, binarization is performed on the detected edges. As a result, unclear information, that is, noise is removed.

In operation S330, an isolated region for eyes, nose, and mouth is generated.

In operation S340, candidate regions for eyes, nose, and mouth are generated by using contrast information or position information of the isolated region. Each of the eye candidate area, the nose candidate area, and the mouth candidate area may be at least one.

In operation S350, an eye region, a nose region, and a mouth region, which are regions where actual eyes, noses, and mouths are present, are extracted from candidate regions for eyes, nose, and mouth through an area minimum / maximization algorithm.

Finally, in step S360, a face image is extracted based on the extracted eye area, nose area, and mouth area.

2. Determination of normal user through facial expression analysis

Since fraudulent users attempt financial transactions without covering the face's major components, even if the face's major components are clearly detected from the face image, the user requesting the financial transaction is not necessarily a normal user. none.

However, the anxiety or anxiety felt by the fraudulent user who attempts illegal financial transactions without covering the main components of the face may be revealed with a facial expression. Therefore, the controller 250 may analyze the facial expressions extracted from the face image, and determine whether the user who requested the financial transaction is a normal user by further considering the analyzed facial expressions.

That is, when an anxiety / fool facial expression, a borderline facial expression, or the like is derived from the analyzed facial expressions, the controller 250 may determine that the user who requested the financial transaction is a fraudulent user.

Hereinafter, an example of the present invention for extracting a facial expression from a face image will be described with reference to FIG. 4.

4 is a diagram illustrating a method of extracting a facial expression from a face image in accordance with an embodiment of the present invention.

In operation S370, geometrical features of an eye region, a nose region, and a mouth region extracted in steps S310 to S360 of FIG. 3 are detected.

Examples of the geometric features include the position of facial components on the face, the degree of opening and closing of the eyes, the degree of opening and closing of the mouth, the distance between the eye and the eyebrows, the degree of wrinkles on the face, and the like.

In operation S380, the facial expression is extracted from the facial image. As an example, in operation S380, the facial expression may be extracted by comparing the predetermined facial expression with the extracted geometrical features.

3. Determination of normal user through analysis of user's behavior pattern

Anxiety or anxiety when the fraudulent user attempts an illegal financial transaction may be revealed not only by the facial expression but also by the fraudulent user's behavior.

In one example, a fraudulent user may indicate anxiety or anxiety by acting around to look around while attempting a financial transaction.

Therefore, the controller 250 analyzes the user's behavior pattern based on the image of the user acquired from the image acquirer 220, and if the analyzed behavior pattern is recognized as an action indicating anxiety or nervousness, requesting a financial transaction. The user can be determined to be a fraudulent user.

4. Determination of whether or not a normal user through the input pattern of the key input received from the key input unit 211

Anxiety or anxiety of the fraudulent user generated through illegal financial transactions is an operation pattern of the financial automation device 200 for preventing illegal financial transactions, that is, input of the key input unit 211 provided in the financial automation device 200 for preventing illegal financial transactions. This can also be confirmed by analyzing the pattern.

As an example, a user attempting a financial transaction may repeat the act of incorrectly entering the passbook or card password several times. This can happen even if the user simply forgets the passbook or card's password, but can also occur if the user is a fraudster. In this case, the controller 250 may determine that the user who attempted the financial transaction is a fraudulent user when the number of wrong input of the password exceeds a preset number.

As mentioned above, even a normal user may forget his or her password, so it may be desirable to analyze the input pattern of the key input together with other judgment factors rather than independently.

5. Determination of normal user through analyzing temperature information

Even if the main components of the face are clearly detected from the face image, and anxiety or anxiety is not revealed in the facial expression, there is a possibility that the user requesting the financial transaction is an illegal user.

For example, when a user exposes a face photograph of another person having a size similar to that of a general person to a camera provided in the image acquisition unit 220, or when the user wears a mask and attempts a financial transaction, the financial transaction Although the requesting user is a fraudulent user, the controller 240 may mistake a user who has requested a financial transaction as a normal user.

When the fraudulent user attempts an illegal financial transaction as in the example above, the face area of the user is covered by a picture or a mask. In general, there may be a difference in the temperature distribution between a face that is not covered by the picture or the mask and a face that is not. .

Therefore, the controller 250 may further determine whether the user who requested the financial transaction is a normal user by further considering temperature information in the face region of the user acquired by the body temperature information acquirer 240.

That is, when the body temperature in the face region is included in the preset temperature distribution range, the controller 250 determines that the user who requested the financial transaction is a normal user, and otherwise determines that the user is a fraudulent user. The range of the preset temperature distribution may be determined based on an average temperature at a general face part of a person.

Hereinafter, the process of obtaining the body temperature information in the face region of the user by the body temperature information acquisition unit 240 will be described in detail with reference to FIG. 5.

5 is a diagram illustrating a method of acquiring body temperature information in a face region of a user according to an embodiment of the present invention.

First, in step S510 collects body temperature information for the user. For example, in step S510, body temperature information of a user may be collected using a thermal infrared camera or a thermal infrared sensor.

In operation S520, the face region of the user is searched based on the collected temperature information of the user. That is, step S520 is a step of searching for a face area from the user's body temperature information.

In general, body temperature information in the face region is uniformly distributed in an ellipse shape at a high temperature of 80 ° F or more. Therefore, in step S520, the temperature information data of 80 ° F or more is accumulated in the vertical and horizontal directions from the temperature information of the user, and the rectangular area in which the face area exists in the vertical and horizontal coordinates of the accumulated area values. Is detected.

Finally, in step S530, when the extracted body temperature information is smaller than the reference value, the body temperature information is extracted from the detected rectangular region, and the user requesting the financial transaction is determined to be a fraudulent user.

6. Determination of normal user through analysis of face material information

As described above, when a fraudulent user exposes a face photo of another person to a camera or attempts to illegally trade a face by wearing a mask, the face image acquired by the image acquirer 220 may be a photo or mask. Since it is obtained by photographing, this is different from the face material of the face image extracted from the face of a general person.

Therefore, the controller 250 analyzes the face material of the face image of the user acquired from the image acquisition unit 220 and compares whether the analyzed face material is similar to the material of the face of a general person. It can be determined whether the user is a normal user.

As an example, the controller 250 may consider the skin information of various races stored in advance, whether the makeup is such as foundation, etc., so that the light reflection value, the light absorption value, the light transmission value of the skin and the subject's face recognition during camera shooting are possible. It is possible to determine whether the photographed face image is a photograph or a mask or a human face by using information such as light transmittance of glasses having a transmittance of.

As mentioned above, the controller 250 determines whether the user is a normal user by appropriately combining the face image, the facial expression, the temperature information, the face material information, the key input pattern, and the user's behavior pattern. Accordingly, the accuracy of the normal user judgment can be improved.

At this time, according to an embodiment of the present invention, when the controller 250 determines that the user who requested the financial transaction is not a normal user, the image acquirer 220 may retake the user to acquire the user image again. have. In this case, the controller 250 notifies the user of the reason for canceling the transaction for the financial transaction request, and the image acquirer 220 may retake the user only when the user agrees to retake the image. Thereafter, the extractor 230, the body temperature information acquirer 240, and the controller 250 re-perform an operation of determining whether the user who requested the financial transaction is a normal user using the re-acquired user image.

For example, the image acquirer 220 may transmit a guide message for inducing a face of the user who requested the financial transaction toward a camera (not shown) provided in the image acquirer 220. In this case, the image acquirer 220 may acquire a clear face image of the user.

In addition, according to an embodiment of the present invention, if it is determined that the user who requested the financial transaction is not a normal user, the controller 250 may transmit a warning message related to the financial transaction request to a financial server (not shown).

As an example, the controller 250 may transmit a warning message to a financial server (not shown) when the number of times the user who has requested the financial transaction is determined to be a fraudulent user exceeds a preset number.

At this time, the financial server (not shown) may notify the holder of the account or card associated with the requested financial transaction based on the transmitted warning message to inform the attempted fraudulent transaction details via e-mail, text message, or telephone. As an example, the financial server may notify the bank account or the cardholder of the fraudulent transaction details by sending a text message as follows.

<The withdrawal transaction attempted at the ATM installed in OOO at xxh xx on xxxxxx, xxxx was not completed. >

Accordingly, the name holder of the bank account or the card can quickly cope with the attempted illegal transaction, thereby preventing the second illegal transaction that may occur later.

In addition, according to an embodiment of the present invention, the financial automation device 200 for preventing illegal financial transactions may further include a storage unit 260 for storing a user image acquired by the image acquisition unit 220. The user image stored in the storage unit 260 may be used to search for a fraudulent user who subsequently attempted an illegal financial transaction.

Hereinafter, a method of controlling the financial automation device 200 for preventing illegal financial transactions described in FIG. 2 will be described in detail with reference to FIG. 6.

6 is a flowchart illustrating a method of controlling a financial automation device for preventing illegal financial transactions according to an embodiment of the present invention.

First, in step S601, a financial transaction request is received from a user.

In operation S602, a user who requests a financial transaction is photographed to obtain a user image of the user.

In operation S603, a face image is extracted from the obtained user image.

In operation S604, a facial expression is extracted from the extracted face image, and in operation S605, facial material information is extracted from the extracted facial image.

In operation S606, body temperature information of the face region of the user is extracted.

In step S607, it is determined whether the user is a normal user based on at least one of a face image, a facial expression, facial material information, and body temperature information.

If it is determined in step S607 that the user is a normal user, step S608 continues the requested financial transaction.

On the contrary, if it is determined in step S607 that the user is a fraudulent user, in step S609 it is determined whether the number of times determined to be a fraudulent user exceeds a preset number.

At this time, if the number of times determined to be the illegal user does not exceed the preset number, the process returns to step S607 and steps S602 to S607 are repeatedly performed.

On the contrary, if the number of times determined to be a fraudulent user exceeds a preset number, in step S610, the requested financial transaction is stopped, and in step S611, a warning message is transmitted to the financial server.

So far, embodiments of the method for controlling the financial automation device for preventing illegal financial transactions according to the present invention have been described, and the configuration of the financial automation device for preventing illegal financial transactions described above with reference to FIGS. 2 to 5 is applied to the present embodiment as it is. It is possible. Hereinafter, a detailed description will be omitted.

In addition, embodiments of the present invention may be implemented in the form of program instructions that may be executed by various computer means to be recorded on a computer readable medium. The computer readable medium may include program instructions, data files, data structures, etc. alone or in combination. Program instructions recorded on the media may be those specially designed and constructed for the purposes of the present invention, or they may be of the kind well-known and available to those having skill in the computer software arts. Examples of computer-readable recording media include magnetic media such as hard disks, floppy disks, and magnetic tape, optical media such as CD-ROMs, DVDs, and magnetic disks, such as floppy disks. Examples of program instructions such as magneto-optical, ROM, RAM, flash memory, etc. may be executed by a computer using an interpreter as well as machine code such as produced by a compiler. Contains high-level language codes. The hardware device described above may be configured to operate as one or more software modules to perform the operations of one embodiment of the present invention, and vice versa.

In the present invention as described above has been described by the specific embodiments, such as specific components and limited embodiments and drawings, but this is provided to help a more general understanding of the present invention, the present invention is not limited to the above embodiments. For those skilled in the art, various modifications and variations are possible from these descriptions. Therefore, the spirit of the present invention should not be limited to the described embodiments, and all of the equivalents or equivalents of the claims as well as the claims to be described later will belong to the scope of the present invention. .

1 is a block diagram showing a general structure of a financial automation device according to an embodiment of the present invention.

2 is a block diagram illustrating an internal configuration of a financial automation device for preventing illegal financial transactions according to an embodiment of the present invention.

3 is a diagram illustrating a method of extracting a face image from a user image step by step according to an embodiment of the present invention.

4 is a diagram illustrating a method of extracting a facial expression from a face image in accordance with an embodiment of the present invention.

5 is a diagram illustrating a method of acquiring body temperature information in a face region of a user according to an embodiment of the present invention.

6 is a flowchart illustrating a method of controlling a financial automation device for preventing illegal financial transactions according to an embodiment of the present invention.

<Explanation of symbols for main parts of the drawings>

200: financial automation devices to prevent illegal financial transactions

210: user interface

211: key input unit

212: card slot

213: bank entry slot

220: image acquisition unit

230: extraction unit

240: body temperature information acquisition unit

250: control unit

260: storage unit

Claims (17)

A user interface for receiving a financial transaction request from a user; An image obtaining unit which photographs the user and obtains a user image of the user; An extraction unit extracting a face image from the user image and extracting a facial expression from the face image; A body temperature information obtaining unit obtaining body temperature information in the face area of the user; And A controller for determining whether the financial transaction request is processed by determining whether the user is a normal user based on the face image, the facial expression, and the temperature information. Financial automation device for preventing illegal financial transactions, including. The method of claim 1, And the extractor extracts geometric features of an eye region, a nose region, and a mouth region from the face image, and extracts the facial expressions by comparing the extracted geometric features with a preset facial expression. The method of claim 1, And the image obtaining unit receives information on a shooting time from the controller when the financial transaction request is received through the user interface, and acquires the user image during the shooting time. The method of claim 1, The user interface A key input unit to receive a key input from the user; And A card slot for receiving a card from the user Including, The image acquisition unit receives illegal information about the shooting time from the control unit when any one of the reception of the key input, the insertion of the card, and the entry of the bankbook occurs, and acquires the user image during the shooting time. Financial automation devices to prevent transactions. The method of claim 1, The extractor further extracts face material information from the face image, And the control unit further determines whether the user is a normal user based on the facial material information. The method of claim 1, The controller analyzes the behavior pattern of the user from the user image, and determines whether the user is a normal user based on the behavior pattern further. The method of claim 1, The user interface is a key input unit for receiving a key input from the user More, And the controller analyzes an input pattern of the key input and determines whether the user is a normal user based on the input pattern. The method of claim 1, If the controller determines that the user is not a normal user, the controller notifies the user of the reason for canceling the transaction for the financial transaction request. And the image obtaining unit re-photographs the user when receiving the consent from the user through the user interface. The method of claim 1, The control unit And if it is determined that the user is not a normal user, transmitting a warning message related to the financial transaction request to a financial server. The method of claim 1, A storage unit storing the user image Financial automation device for preventing illegal financial transactions further comprising. A user interface having a key input unit for receiving a key input from a user, the user interface receiving a financial transaction request from the user; An image obtaining unit which photographs the user and obtains a user image of the user; An extraction unit extracting a face image from the user image and extracting a facial expression from the face image; And A controller for determining whether the financial transaction request is processed by determining whether the user is a normal user based on the face image, the facial expression, and the input pattern of the key input Financial automation device for preventing illegal financial transactions, including. The method of claim 11, The extractor further extracts face material information from the face image, And the control unit further determines whether the user is a normal user based on the facial material information. The method of claim 11, The controller analyzes the behavior pattern of the user from the user image, and determines whether the user is a normal user based on the behavior pattern further. The method of claim 11, The control unit And if it is determined that the user is not a normal user, transmitting a warning message related to the financial transaction request to a financial server. In a method for controlling a financial automation device for preventing illegal financial transactions, Receiving a financial transaction request from a user; Photographing the user to obtain a user image for the user; Extracting a face image from the user image; Extracting a facial expression from the face image; Extracting body temperature information in the face area of the user; Determining whether the user is a normal user based on the face image, the facial expression, and the temperature information; And Determining whether to process the financial transaction request according to whether the user is a normal user Control method of a financial automation device for preventing illegal financial transactions comprising a. The method of claim 15, Receiving the financial transaction request includes receiving a plurality of user inputs from the user, Acquiring the user image When a first user input is received among the plurality of user inputs, photograph the user to obtain a first user image, In the step of extracting the face image, if the face image cannot be extracted from the first user image, control of the financial automation device for preventing illegal financial transactions to retake the user to obtain a second user image for the user Way. The method of claim 16, Acquiring the user image If the face image can be extracted from the first user image in the step of extracting the face image, the control method of the financial automation device for preventing illegal financial transactions to stop shooting of the user.

Images