KR20210005332A - Apparatus and method for the guidance of elderly persons in smart automated teller machine using facial recognition - Google Patents

Abstract

The present invention relates to a device and method for inducing a senior citizen to video consultation of a smart automatic teller machine using face recognition, which recognizes a senior citizen using a face recognition technology in a smart automated teller machine (hereinafter, referred to as ″STM″), and induces the recognized senior citizen to video consultation to allow convenient use of an STM. The method of the present invention comprises the steps of: transmitting a preset basic guide voice when a user access is sensed for a predetermined time or more; executing a face recognition solution for user recognition, photographing a user face by obtaining a control authority of a face recognition camera, and estimating a user age through analysis of a photographed user face image; returning the control authority of a camera when estimation of the user age is completed, ending a face recognition process, and transmitting a face recognition result to a financial business control unit; and inducing video consultation by interworking with a video consultation connecting inducing unit for inducing video consultation connection when the estimated user age is determined as a senior citizen who has low capability in digital informatization. Therefore, even the senior citizen who has low capability in digital informatization can conveniently use the STM.

Description

Apparatus and method for the guidance of elderly persons in smart automated teller machine using facial recognition}

The present invention recognizes an elderly person by using facial recognition technology in a smart automated teller machine (hereinafter referred to as "STM"), and induces video counseling to the recognized elderly person to conveniently use STM. It relates to an apparatus and method for inducing video counseling for the elderly in a smart automated financial device using recognition.

According to Digital Transformation, counter-in-counter services are performed only through a bank teller, and non-face-to-face business through automated financial devices installed in the bank. It is being done. Here, as the mandatory authentication method for non-face-to-face real name verification, the real name is verified through at least two of: submission of a copy of the ID card, video call, confirmation when delivering an access medium (real transaction confirmation), account utilization (designated account transfer), and biometric authentication. Confirm.

In recent years, in the financial market, smart financial automation devices (STMs) that are capable of counter work and automated device (ATM) work are spreading according to Digital Transformation.

Techniques previously proposed for automated financial devices are disclosed in <Patent Document 1> and <Patent Document 2> below.

The prior art disclosed in <Patent Document 1> is a main body having a medium input/output unit for depositing and withdrawing media, a display unit provided in the main unit, and a display unit that selectively displays a repair selection unit for activating a repair service function, and a repair service. And a storage box in which an article for receiving support is stored, a communication unit communicatively connected to a service terminal for supporting a repair service for the stored article, and a control unit for opening the storage box when the repair selection unit is selected on the display unit. When it is confirmed that the article is stored in the storage box, the control unit transmits the storage information on the article to the service terminal. Through this configuration, it is possible to perform a repair service function in addition to the financial transaction service in the automated financial device.

In addition, the prior art disclosed in <Patent Document 2> is a step of receiving check card application information including customer information and storing it in a temporary database, receiving check card application information from a customer terminal, and responding to check card application information. Displaying a check card application page for applying for a check card on the screen, when the customer's account number and password are entered through the check card application page, generating a check card application based on the customer information, account number and password And when a signature is input to the check card application, after completing the application for the check card based on the check card application, transmitting an order for issuing the check card to the automated financial device. Through this configuration, a check card is issued using an automated financial device.

Republic of Korea Patent Publication 10-2019-0020358 (published on Mar. 4, 2019) (Financial devices, control methods and financial systems of financial devices) Republic of Korea Patent Publication 10-2014-0000509 (published on Jan. 3, 2014) (debit card issuing method and system using automated financial devices)

However, the general smart automated financial device and the automated financial device mentioned in the patent literature as described above are available for repair service or check card application through the counter in addition to the normal financial automation work, but the elderly with low digital informatization capabilities can It is more difficult to use than to handle it face-to-face with the staff.

In other words, the use of smart automated financial devices can be used by all age groups, but it is difficult for an elderly person with low digital information capabilities to directly manipulate many menus that enable them to handle counter/ATM tasks.

This difficulty in use consequently acts as an obstacle to the popularization of smart automated financial devices, and acts as a factor that hinders the spread of smart automated financial devices.

Therefore, the present invention has been proposed to solve the problems arising from the above-described general smart automated financial device and the prior art, and recognizes the elderly using facial recognition technology in the smart automated financial device, and provides image consultation to the recognized elderly. The purpose of this is to provide a device and method for inducing video counseling for the elderly using a smart financial automation device using facial recognition so that even elderly people with low level of digital information capability can use STM conveniently.

In order to achieve the above object, the apparatus for inducing image counseling for the elderly of the smart automated financial device according to the present invention comprises: a main body; A user detection unit provided in the main body and detecting a user's approach; A camera unit provided in the main body and photographing an image for video consultation or user recognition; A user recognition unit that recognizes a user's facial association photographed by the camera unit and outputs age group estimation information; A video consultation connection inducing unit for inducing a video consultation connection when it is determined as an elderly person having a low digital information capability based on the age estimation information output by the user recognition unit; A financial business control unit that controls the camera unit to capture a user's facial image when a user access is detected by the user detection unit, and controls the induction of a video consultation connection in connection with the video consultation connection guidance unit; And a customer manipulation unit for outputting a screen for inducing video consultation under the control of the financial business control unit and transmitting operation information of the user to the financial business control unit.

In the above, the camera unit includes a video consultation camera for photographing an image of a user during video consultation; It characterized in that it comprises a face recognition camera for photographing the user's face to estimate the user age group.

In the above, the video counseling connection inducing unit is characterized in that when it is determined as an elderly person based on the age group estimation information, it generates screen information that induces the video counseling connection and displays it to the customer control unit through the financial affairs control unit to induce video counseling.

In the above, the financial affairs control unit ignores the user recognition result of the user recognition unit when a user recognition end situation occurs while performing the facial recognition solution for estimating the user's age according to the user detection, and switches to the initial screen or finances according to the user input. It is characterized by conversion to work.

In the above, the end state of user recognition is characterized in that it is determined as the end state of user recognition when a user's input occurs prior to the estimation of the user's age, the user is not detected, or the user recognized by the face recognition camera is changed.

In order to achieve the above object, the method for inducing image counseling for the elderly in a smart automated financial device according to the present invention includes: (a) When a user's access is detected by a user detection unit built in the smart automated financial machine for a predetermined time or longer, Transmitting a set basic guide voice; (b) executing a face recognition solution for user recognition in the user recognition unit, obtaining control rights of the face recognition camera to photograph the user's face, and estimating the user age through analysis of the captured user face image; (c) when the estimation of the user's age range is completed, returning the camera control right, ending the facial recognition process, and transmitting the facial recognition result to the financial affairs control unit; (d) inducing video counseling by linking with a video counseling connection inducing unit that induces a video counseling connection when the estimated age of the user is determined to be an elderly person; (e) As a result of inducing video counseling in step (d), if the user selects video counseling, the financial business control unit controls the video counseling camera and the customer control unit to conduct financial business through video counseling, and the user performs video counseling. If not selected, it characterized in that it comprises the step of switching to the general financial business progress screen.

In the above, step (b) includes the steps of: (b1) executing a face recognition solution and obtaining camera control rights; (b2) estimating the age of the user by analyzing the image acquired through the facial recognition camera; (b3) returning the camera control right when estimation of the user's age range is completed; (b4) transmitting the result of facial recognition to the financial business control unit,

In step (b4), if the face recognition result is not a human face, if face recognition is not performed for a predetermined period of time or longer, or if a camera failure occurs, the face recognition result is output as a face recognition error. It is characterized in that the user age group estimation information is output as a result of facial recognition.

In the above step (d), when it is determined as an elderly person based on age group estimation information, screen information for inducing video consultation connection is generated and displayed on the customer control unit through the financial business control unit to induce video consultation.

In addition, the method of inducing image counseling for the elderly in a smart automated financial device according to the present invention includes: (f) ignoring the user recognition result of the user recognition unit when a user recognition end situation occurs while the user's facial image is recognized through the face recognition solution. , It characterized in that it further comprises the step of switching to the initial screen or switching to the financial business according to the user input.

In the above, the end state of user recognition is determined as the end state of user recognition when a user input occurs prior to the estimation of the user's age, the user is not detected for more than a certain period of time, or the user recognized by the face recognition camera is changed, and the user input When this occurs, it switches to the financial business corresponding to the input, and when the user is not detected for a certain period of time or when the user is changed, the initial screen is switched to improve security.

According to the present invention, the smart financial automation device recognizes the elderly using facial recognition technology and induces video consultation to the recognized elderly, thereby promoting the convenience of using STM even for the elderly with low level of digital information capability. have.

In addition, according to the present invention, by allowing everyone of all ages to conveniently use STM, there is also an advantage of contributing to the popularization and spread of STM.

1 is a schematic configuration diagram of a device for inducing image counseling for an elderly person of a smart automated financial device using facial recognition according to the present invention;
FIG. 2 is a block diagram showing an embodiment of a device for inducing image counseling for an elderly person of a smart automated financial device using facial recognition according to the present invention;
3 is a flow chart showing a method of inducing image counseling for the elderly using a smart automated financial device using facial recognition according to the present invention;
4 is an exemplary diagram illustrating a video counseling guidance screen in the present invention.

Hereinafter, an apparatus and method for inducing image counseling for the elderly in a smart automated financial device using facial recognition according to a preferred embodiment of the present invention will be described in detail.

The terms or words used in the present invention described below should not be construed as being limited to a conventional or dictionary meaning, and the inventor may appropriately define the concept of terms in order to describe his own invention in the best way. It should be interpreted as a meaning and concept consistent with the technical idea of the present invention based on the principle that there is.

Therefore, the embodiments described in the present specification and the configurations shown in the drawings are only preferred embodiments of the present invention, and do not represent all the technical spirit of the present invention, and various equivalents that can replace them at the time of the present application It should be understood that there may be variations.

<Example 1>

1 is a schematic configuration diagram of a device for inducing image counseling for an elderly person of a smart automated financial device using facial recognition according to a preferred embodiment of the present invention, and FIG. 2 is a block diagram of a device for inducing image counseling for an elderly person using facial recognition. It is a configuration diagram.

Herein, the term "elderly person" means a person over a preset age, and the criteria for determining the age of an elderly person may be different, but it is assumed that a person aged 60 or older is referred to as an elderly person.

The image counseling guidance device for the elderly of the smart financial automation device (STM) using facial recognition according to the present invention has a built-in camera for image counseling, facial recognition, etc., tasks requiring authentication (also referred to as counter service) and financial automation tasks A main body 120 in which various devices for performing the operation are embedded is provided, and the main body 120 is provided with a user detection unit 101 that detects a user's approach.

Here, the tasks that require authentication include opening a bankbook and verifying biometric information, and the financial automation work includes transactions such as cash deposit and withdrawal and account transfer.

Such a user detection unit 101 may use a proximity sensor to detect that the user approaches the main body 120 of the smart financial automation device 100, and a known proximity sensor capable of detecting user access other than this proximity sensor Various detection sensors can also be used.

The main body 120 is provided with a camera unit 102 for capturing an image for video consultation or user recognition. Here, the camera unit 101 is provided with an image counseling camera 1021a for image counseling and a face recognition camera 102b for user face recognition.

In addition, in the main body 120, a user recognition unit 103 that recognizes a user's facial association photographed by the camera unit 102 and outputs age group estimation information, and age group estimation information output by the user recognition unit 103 If it is determined that an elderly person with a low digital information capability is determined, a video consultation connection inducing unit 111 that induces a video consultation connection, and when a user access is detected by the user detection unit 101, the camera unit 102 is controlled to A financial business control unit 105 is provided to capture a facial image, control the induction of video counseling connection by interlocking with the image counseling connection induction unit 111, and integrally control the counter service and the financial automation business.

In addition, the main body 120 outputs a screen for inducing video consultation under the control of the financial business control unit 105, and a customer operation unit 104 that transmits the user's operation information to the financial business control unit 105 It is equipped. The customer manipulation unit 104 displays information and inputs customer information using a single touch display.

In addition, the main body 120 includes a voice memory 108 in which a voice or control program to be provided to a user is embedded, an identification card recognition unit 109 for recognizing an identification card presented by the user, and a biometric recognition unit for recognizing the user's biometric information. (110), a voice transmission unit 106 for transmitting the voice signal output from the financial service control unit 105 through the speaker 107 is provided.

In addition, the main body 120 includes a function for recognizing information (passbook, etc.) necessary for a counter service or financial business requiring authentication in a conventional smart automated financial device or organizing a bankbook.

The operation of the image counseling induction device for the elderly of the smart financial automation device (STM) using facial recognition according to the present invention configured as described above will be described in detail as follows.

First, when the user approaches the main body 120 within a certain distance while the smart automatic financial device 100 is waiting for user detection, the proximity sensor of the user detection unit 101 detects it and detects the user. An event (user detection signal) is generated and transmitted to the financial affairs control unit 105.

When an event for user detection occurs, the financial business control unit 105 switches from the standby mode to the financial business mode, and transmits a pre-registered basic customer guidance voice through the voice transmission unit 106. At the same time, the user input is monitored through the financial customer manipulation unit 104 and at the same time, the face recognition camera 102b of the camera unit 102 is activated to capture the user's face. The facial recognition camera 102b is activated to capture the user's face, and transmits the captured user's facial image to the user recognition unit 103. Here, the facial recognition camera 102b is preferably installed in a position capable of photographing all the faces of a disabled or normal person using a wheelchair.

The user recognition unit 103 executes a facial recognition solution, analyzes the transmitted user facial image, and estimates a user age group (age). In the method of estimating the age group by analyzing the photographed user's face image, a known technique for estimating the age group through the existing face image may be used as it is.

For example, a face recognition service method disclosed in Korean Patent Registration No. 10-1835333 (name of invention: face recognition service provision method) may be used.

That is, an image of a facial image is input, a face is searched from the input image, and facial feature points are extracted by classifying eyes, nose, mouth, and chin from the searched face. Thereafter, an aging feature point having a similarity to a reference aging feature point previously matched to the location of the extracted facial feature point and stored above a preset reference value is searched for from the image. Next, after reading the meta data of the searched user aging feature points and the reference aging feature points that are equal to or greater than a preset reference value, the user's age group is estimated using the read meta data.

The user age group estimation information estimated in this way is transmitted to the financial affairs control unit 105.

The financial affairs control unit 105 checks the user age group estimation information estimated by the user recognition unit 103, and when it is determined that it is an elderly person with a low level of digital informatization capability (for example, 60 years of age or older), video consultation is connected. Interworking with the induction unit 111 to induce video counseling.

The video counseling connection inducing unit 111 generates screen information for inducing video counseling connection when the user age group estimated by the financial business control unit 105 is an elderly person with low digital information capability, and the financial business control unit 105 Through a pop-up form on the customer control unit 104 is expressed.

4 is an example of an image counseling connection guidance screen displayed on the customer operation unit 104. The video consultation screen is displayed, and the user (customer) selects whether to connect to the video consultation. The user can select whether to connect to the video consultation by touching "Yes" or "No" on the customer screen. If necessary, it is possible to display a video counseling connection guidance screen on the customer control unit 104 and simultaneously transmit a voice for the video counseling connection through the voice transmission unit 106.

At this time, if the user selects "No", the screen is converted to the basic financial business screen, and financial business corresponding to the user's input is performed. On the contrary, if the user selects "Yes", it switches to the video consultation mode, and After activating the counseling camera 102a, a bank employee at the counter is connected to conduct financial business in the same customer service manner as in which the bank conducts financial business face-to-face.

On the other hand, as another feature of the present invention, if the user recognition termination situation occurs while the user's age group estimation is not completed while performing the facial recognition solution for estimating the user's age according to the user detection, The user recognition result of the user recognition unit 103 is ignored, and the screen is switched to an initial screen or a financial service is switched to a user input.

Here, the end of user recognition is determined by the user's manipulation of the customer manipulation unit 104, card input or passbook input, etc., prior to the estimation of the user's age, or the user is not detected, or is recognized by the face recognition camera. If the changed user is changed, it is determined as the end of user recognition. Here, the method of determining the situation where the user is not detected is to perform the user detection operation continuously for a predetermined time if the user is not detected in the situation where the user is detected, and quickly recognize the user if the user is not detected during the predetermined time. It judges it as the end situation, ends the current task, and switches to the initial screen. In addition, if the user changes during the continuous user detection and recognition operation in a situation where the user is detected, it is quickly determined as the user recognition end situation, ends the current task, and switches to the initial screen. Whether or not the user has been changed can be easily checked whether or not the user has changed by continuously comparing the previous user face recognition information with the user's face recognition information recognized in real time.

For example, the user recognition unit 103, while executing the facial recognition solution under the control of the financial business control unit 105, when a plurality of user's face images are recognized, the plurality of users through the financial business control unit 105 Notifies that a face image is being recognized.

When the financial affairs control unit 105 recognizes that a plurality of user face images are recognized during facial image recognition, a plurality of facial images are recognized through the customer control unit 104, so that only one person who actually makes a financial transaction remains. Send out. After that, the timer is activated, and if only one person's face image remains within the set time (e.g., 5 seconds), the financial transaction continues. Unlike this, if multiple face images are detected after the set time, automatic Log out forcibly and return to the initial screen.

On the other hand, if only one person's face image remains within the set time, the corresponding face image is registered as a user image, and user face recognition monitoring is started.

The user image recognized through the facial recognition solution is registered and the user image that is continuously acquired until the financial transaction is terminated is analyzed to monitor the user's departure and change. When registering a user image, an ID can be assigned to register. Here, as a method of monitoring user departure and change, user change is monitored through facial tracking technology that extracts a face image from the image acquired from the face recognition camera 102b in real time and compares it with the registered user image to check the degree of consistency. do. In addition, through image analysis acquired from the facial recognition camera 102b, if a face image is not recognized, or if only an image other than a face is recognized, it is determined that the user has deviated, and a face image recognized through the face tracking technology is registered. If it is different from the face image, it is determined that the user has changed. In addition, when a user departure or change is detected, a detection event for the user departure or change is transmitted to the financial affairs control unit 105.

When a user departure or change detection event occurs from the user recognition unit 103 during a financial transaction, the financial affairs control unit 105 checks whether the generated event is a user departure event. Here, by assigning different IDs to each of the user exit event and the user change event, the financial affairs control unit 105 can easily determine whether the user exits or the user is changed.

As a result of analyzing the event information, the financial affairs control unit 105, if a user departure detection event signal, broadcasts an announcement about user departure through the voice transmission unit 106 (for example, if a financial transaction is desired, it moves to a designated seat) An announcement that recommends). During an actual financial transaction, a user may drop belongings or make a phone call, and in this case, there may be cases where the user falls away from the smart automatic financial device or the location of the face recognition camera. In this case, if the user proceeds to log out immediately, there is an inconvenience in that the user has to proceed with the financial business from the beginning. To improve this inconvenience, when a user churn detection event occurs, a timer is activated, and if the user returns within a preset time (for example, 10 seconds), the current financial transaction is maintained so that the user can initiate the financial transaction from the beginning. It relieves the inconvenience of starting over. On the contrary, if the user does not return within a preset time, the user automatically logs out to protect personal information and switches to the initial screen to prevent leakage of personal information, thereby enhancing security.

In contrast, when a user change event occurs during a financial transaction, the financial affairs control unit 105 automatically logs out and switches to the initial screen to protect personal information to prevent personal information leakage and enhance security.

As described above, if the user is not detected or the user is changed, the current task is promptly terminated and initialized, thereby preventing others from checking the user's personal information through the current screen or returning to the previous screen to steal the user's personal information. Can be prevented.

In addition, the financial affairs control unit 105 checks the user face recognition result transmitted from the user recognition unit 103, and analyzes that it is not a human face, or is not detected for more than a certain period of time, or a human face is recognized due to a camera failure. When the result of analysis indicating that the user did not lose is delivered, the task of inducing user recognition and video consultation connection is terminated.

<Example 2>

3 is a flow chart showing a method of inducing video counseling for the elderly by the smart automated financial device according to the present invention, (a) the user's access is detected by the user detection unit 101 built in the smart automated financial device 100 for a predetermined time or longer. Then, the step of transmitting a preset basic guide voice (S101-S102), (b) the user recognition unit 103 executes a face recognition solution for user recognition, and acquires the control right of the face recognition camera 102b. Photographing the face of the user and estimating the user's age through the analysis of the captured user's facial image (S103-S106), (c) when the estimation of the user's age range is completed, the camera control right is returned and the facial recognition process is terminated, The step of delivering the facial recognition result to the financial business control unit 105 (S107-S108), (d) a video counseling connection inducing unit that induces a video counseling connection when the age range of the user estimated by the financial business control unit 105 is determined to be an elderly person Steps (S111-S112) inducing video counseling in connection with (111), (e) As a result of inducing video counseling in step (d), when the user selects video counseling, the financial affairs control unit 105 It includes a step (S113) of controlling (102a) and the customer operation unit 104 to perform financial business through video consultation, and if the user does not select the video consultation, switching to a general financial service progress screen (S113).

The step (b) includes the steps of (b1) executing a face recognition solution and obtaining camera control rights (S104-S105), (b2) estimating the age of the user by analyzing the image acquired through the face recognition camera (S106). ), (b3) returning the camera control right when estimation of the user's age group is completed (S107), (b4) transferring the facial recognition result to the financial affairs control unit 105 (S108), and the (b4) Steps are: If the face recognition result is not a human face, if face recognition is not performed for a predetermined period of time or longer, or if a camera failure occurs, the face recognition result is output as a face recognition error.If the user age group is estimated, the user age group estimation information Is output as the result of facial recognition.

In addition, the method of inducing image counseling for the elderly by the smart automated financial device according to the present invention includes (f) the user recognition of the user recognition unit 103 when a user recognition end situation occurs while the user's facial image is recognized through the face recognition solution. The step of ignoring the result and switching to an initial screen or switching to a financial service according to a user input (S114 to S116) is further included.

A detailed description will be given of a method of inducing image counseling for the elderly using a smart financial automation device (STM) using facial recognition according to the present invention configured as described above.

First, in step S101, when the user approaches the main body 120 within a certain distance while the financial affairs control unit 105 of the smart financial automation device 100 waits for user detection, the proximity of the user detection unit 101 The sensor detects this, generates an event (user detection signal) for user detection, and transmits it to the financial affairs control unit 105. In this case, it is preferable to generate the user detection event only when the user is detected for a predetermined period of time or longer.

When an event for user detection occurs, the financial affairs control unit 105 moves to step S102 and transmits a pre-registered basic customer guidance voice through the voice transmission unit 106. Thereafter, the process moves to step S103 to monitor the user's input through the financial customer manipulation unit 104 and at the same time activate the face recognition camera 102b of the camera unit 102 to capture the user's face. The facial recognition camera 102b is activated to capture the user's face, and transmits the captured user's facial image to the user recognition unit 103. Here, the facial recognition camera 102b is preferably installed in a position capable of photographing all the faces of a disabled or normal person using a wheelchair.

The user recognition unit 103 executes the facial recognition solution in step S110, analyzes the transmitted user facial image, and estimates the user age group (age). In the method of estimating the age group by analyzing the photographed user's face image, a known technique for estimating the age group through the existing face image may be used as it is.

For example, as is well known, the face recognition service method disclosed in Korean Patent Registration No. 10-1835333 (name of the invention: face recognition service provision method) may be used.

That is, in step S104, face recognition is started using the face recognition solution, the camera control right is obtained in step S105, and the image acquired through the face recognition camera is moved to step S106 and disclosed in the published face recognition service method. Analyze by face recognition method to estimate user age. Next, the process moves to step S107, and when the user age range estimation is completed, the camera control right is returned, and in step S108, the facial recognition result is transmitted to the financial affairs control unit 105, and the facial recognition solution is terminated.

Here, if the face recognition result is not a human face, if face recognition is not performed for a predetermined period of time or more, or if a camera failure occurs, the face recognition result is output as a face recognition error.If the user age group is estimated, the user age group estimation information Is output as the result of facial recognition.

In step S111, the financial affairs control unit 105 checks the user age group estimation information estimated by the user recognition unit 103, and when it is determined that it is an elderly person with a low level of digital informatization capability (for example, 60 years or older), the step Move to S112 and interlock with the video consultation connection inducing unit 111 to induce video consultation. The video counseling connection inducing unit 111 generates screen information for inducing video counseling connection when the user age group estimated by the financial business control unit 105 is an elderly person with low digital information capability, and the financial business control unit 105 Through a pop-up form on the customer control unit 104 is expressed. 4 is an example of an image counseling connection guidance screen displayed on the customer operation unit 104. The video consultation screen is displayed, and the user (customer) selects whether to connect to the video consultation. The user can select whether to connect to the video consultation by touching Yes or No on the customer screen. If necessary, it is possible to display a video counseling connection guidance screen on the customer control unit 104 and simultaneously transmit a voice for the video counseling connection through the voice transmission unit 106. At this time, if the user selects No, the screen is converted to the basic financial business screen, and financial business corresponding to the user's input is performed. On the contrary, if the user selects Yes, the video consultation mode is switched, and the video consultation camera 102a ), and then, by connecting the bank staff at the counter, the bank will conduct financial business in the same customer-facing manner as in the face-to-face financial business.

On the other hand, as another feature of the present invention, the financial business control unit 105 terminates user recognition as in step S114 in a state in which estimation of the user age group is not completed while performing the facial recognition solution for estimating the user age according to the user detection. When a situation arises, the process moves to step S115, ignoring the user recognition result of the user recognition unit 103, and moves to step S16 to switch to the initial screen or switch to financial business according to the user input.

Here, the end of user recognition is determined by the user's manipulation of the customer manipulation unit 104, card input or passbook input, etc., prior to the estimation of the user's age, or the user is not detected, or is recognized by the face recognition camera. If the changed user is changed, it is determined as the end of user recognition. Here, the method of determining the situation where the user is not detected is to perform the user detection operation continuously for a predetermined time if the user is not detected in the situation where the user is detected, and quickly recognize the user if the user is not detected during the predetermined time. It judges it as the end situation, ends the current task, and switches to the initial screen. In addition, if the user changes during the continuous user detection and recognition operation in a situation where the user is detected, it is quickly determined as the user recognition end situation, ends the current task, and switches to the initial screen. Whether or not the user has been changed can be easily checked whether or not the user has changed by continuously comparing the previous user face recognition information with the user's face recognition information recognized in real time.

As described above, if the user is not detected or the user is changed, the current task is promptly terminated and initialized, thereby preventing others from checking the user's personal information through the current screen or returning to the previous screen to steal the user's personal information. To improve security.

In addition, the financial affairs control unit 105 checks the user face recognition result transmitted from the user recognition unit 103, and analyzes that it is not a human face, or is not detected for more than a certain period of time, or a human face is recognized due to a camera failure. When the result of analysis indicating that the user did not lose is delivered, the task of inducing user recognition and video consultation connection is terminated.

As described above, the present invention estimates the age group through user facial recognition, and if the estimated age group is an elderly person with low digital informatization capability, it induces a video consultation connection, which is the same as conducting financial business face-to-face at the counter. There is an advantage in that it allows the financial business to be performed conveniently, so that even the elderly with low digital information capability can conveniently use the Smart Financial Automatic Machine (STM).

Although the invention made by the present inventor has been described in detail according to the above embodiment, the present invention is not limited to the above embodiment, and it is common knowledge in the art that various changes can be made without departing from the gist of the invention. It is self-evident to those who have

100: smart automatic financial device
101: user detection unit
102: camera unit
102a: video consultation camera
102b: face recognition camera
103: user recognition unit
104: customer control panel
105: financial affairs control unit
106: voice transmission unit
107: speaker
108: voice memory
109: identification card recognition unit
110: biometric recognition unit
111: video counseling connection guidance unit
120: main body

Claims (10)

main body;
A user detection unit provided in the main body and detecting a user's approach;
A camera unit provided in the main body and photographing an image for video consultation or user recognition;
A user recognition unit that recognizes a user's facial association photographed by the camera unit and outputs age group estimation information;
A video counseling connection inducing unit for inducing a video counseling connection when it is determined as an elderly person based on the age estimation information output by the user recognition unit;
A financial business control unit that controls the camera unit to capture a user's facial image when a user access is detected by the user detection unit, and controls the induction of a video consultation connection in connection with the video consultation connection guidance unit; And
And a customer manipulation unit that outputs a screen for inducing video consultation under control of the financial business control unit and transmits the user's operation information to the financial business control unit.
The apparatus of claim 1, wherein the camera unit comprises: an image counseling camera for photographing a user's image during image counseling; A device for inducing image counseling for the elderly of a smart automated financial device, comprising a facial recognition camera that photographs a user's face for estimating the user's age group.
The method of claim 1, wherein the video counseling connection inducing unit generates screen information that induces video counseling connection when it is determined as an elderly person based on age group estimation information and displays it to the customer control unit through the financial affairs control unit to induce video counseling. A device for inducing video counseling for the elderly in a smart automated financial device.
The method of claim 1, wherein the financial business control unit ignores the user recognition result of the user recognition unit and switches to the initial screen or enters a user input when a user recognition end situation occurs while performing the facial recognition solution for estimating the user's age according to the user detection. A device for inducing video counseling for the elderly in a smart automated financial device, characterized in that switching to financial services according to.
In claim 4, wherein the end state of user recognition is determined as the end state of user recognition when a user's input occurs prior to estimation of the user's age, the user is not detected, or the user recognized by the face recognition camera is changed. A device for inducing video counseling for the elderly in a smart automated financial device.
(a) transmitting a preset basic guide voice when the user's access is detected for a predetermined time or longer by a user detection unit built in the smart financial automatic machine;
(b) executing a face recognition solution for user recognition in the user recognition unit, obtaining control rights of the face recognition camera to photograph the user's face, and estimating the user age through analysis of the captured user face image;
(c) when the estimation of the user's age range is completed, returning the camera control right, ending the facial recognition process, and transmitting the facial recognition result to the financial affairs control unit;
(d) inducing video counseling by linking with a video counseling connection inducing unit that induces a video counseling connection when the estimated age of the user is determined to be an elderly person; And
(e) As a result of inducing video counseling in step (d), if the user selects video counseling, the financial business control unit controls the video counseling camera and the customer control unit to conduct financial business through video counseling, and the user performs video counseling. If not selected, the method of inducing video counseling for the elderly in a smart automated financial device, comprising the step of switching to a general financial business progress screen.
In claim 6, the step (b) comprises the steps of: (b1) executing a face recognition solution and obtaining camera control rights; (b2) estimating the age of the user by analyzing the image acquired through the facial recognition camera; (b3) returning the camera control right when estimation of the user's age range is completed; (b4) transmitting the result of facial recognition to the financial business control unit,
In step (b4), if the face recognition result is not a human face, if face recognition is not performed for a predetermined period of time or longer, or if a camera failure occurs, the face recognition result is output as a face recognition error. A method of inducing video counseling for the elderly in a smart automated financial device, characterized in that outputting user age estimation information as a result of facial recognition.
In claim 6, the step (d) is characterized in that when it is determined as an elderly person based on age group estimation information, screen information for inducing video consultation connection is generated and displayed on the customer control unit through the financial affairs control unit to induce video consultation. How to induce video counseling for the elderly using smart automated financial devices.
In claim 6, (f) If a user recognition end situation occurs while the user's facial image is recognized through the face recognition solution, the user recognition result of the user recognition unit is ignored, and the screen is switched to the initial screen or to the financial service according to the user input The method of inducing video counseling for the elderly by a smart automated financial device, characterized in that it further comprises the step of:
In claim 9, wherein the user recognition end situation is determined as a user recognition end situation when a user input occurs prior to estimation of the user's age, the user is not detected for more than a certain time, or the user recognized by the face recognition camera is changed, When the user input occurs, it switches to the financial service corresponding to the input, and when the user is not detected for a certain period of time or when the user is changed, the screen is switched to the initial screen to improve security. How to induce.

Images