KR102244270B1 - HMI for enhancing recognition rate for iris and speed - Google Patents

Abstract

The present invention relates to a human-machine interface (HMI) device in which a controller is configured to extract an iris region corresponding to a distance (Δ) from a user detected by a distance sensor and extract an iris image from the extracted iris region so that an iris recognition rate is increased and acquisition of the iris image is correctly and quickly performed to increase the overall authentication speed and reduce unnecessary time consumption required for authentication. Since the controller can recognize the iris without making the user move the iris to a position close to a camera, an iris recognition process for the user is facilitated, thereby increasing user convenience. Moreover, since an iris authentication system is applied to the HMI device, the user authentication is performed by comparing and contrasting feature patterns of the iris image of the user with feature patterns of previously registered users when the corresponding user approaches, operation and access are permitted to only authorized users, thereby significantly increasing security. The HMI device comprises the distance sensor, a capturing means, and the controller.

Description

HMI device with improved iris recognition rate and authentication speed {HMI for enhancing recognition rate for iris and speed}

The present invention relates to an HMI device with increased iris recognition rate and authentication speed, and in detail, it is configured to automatically extract an iris region according to a distance from a user who is subject to iris authentication, so that the iris image extraction can be performed quickly and accurately. It relates to an HMI device capable of maximizing the authentication speed compared to the same configuration by optimizing the inquiry order of registered feature data of registered users at each period (T).

In general, in various industrial fields along with machine automation, PLC (Programer logic controller) devices that control input/output, operation and power of production lines or mechanical devices are commercialized and used, and digital data of PLC devices along with the development of PLC devices Various studies have been conducted on a human machine interface (HMI) that converts data into human-recognizable data, and preferably displays data on a display device such as a touch screen panel.

These HMIs can easily monitor and control industrial devices in various industrial fields by requesting an operation command for a specific device input from the user and receiving response data corresponding to the requested data value from the device and displaying it. It has the advantage of being able to do it.

In other words, industrial managers can not only monitor the status of various industrial devices through HMI, but also control various industrial devices, so that users who are inexperienced with HMI or who do not have specialized knowledge about industrial devices can In the case of manipulation, it may cause process accidents, failures, and breakdowns. Therefore, an authentication system has been introduced and used that permits manipulation only to authorized users.

On the other hand, as the importance of security has emerged along with informatization in recent years, and with the development of biometrics and electronic device industries, a user authentication system that grants access rights to specific systems or devices only to biometrically authenticated users has been developed. It is widely used in various places and purposes.

In particular, the iris pattern is formed before the age of 3 years of age, and once created, it has a characteristic that hardly changes throughout life.Therefore, there are various studies on the iris recognition system to authenticate users using the iris, which is a unique physiological characteristic of humans. Is progressing.

However, there is still insufficient research to improve security by applying an iris authentication system to an HMI device.

1 is a flow chart showing the HMI program control method disclosed in Korean Patent Publication No. 10-2018-0107548 (name of the invention: HMI program control method through a mobile device).

In the HMI program control method (hereinafter referred to as the prior art) of FIG. 1 (S900), a service request is received from a user and the type of the requested service is checked at step 910 (S910) and the service type identified at step 910 (S910). Step 920 (S920) of determining whether is a general level or a security level, and step 930 of executing a biometric module using a camera mounted on a mobile device and proceeds when the service type identified in step 920 (S920) is a security level. In step 940 (S940) and step 940 (S940) of performing user authentication using the biometrics module executed through step 930 (S930) and step 940 (S940), when authentication fails, the HMI program is deactivated. It consists of step 950 (S950) and step 960 (S960) of displaying response data for the requested service through the mobile device, which is processed when authentication is successful in step 940 (S940).

The prior art (S900) configured in this way performs user authentication through biometrics using a camera, and displays response data for the requested service according to the user authentication result, thereby improving the security of control of the HMI program. In addition, it has the advantage of preventing a malfunction of the power system under the control of an unauthorized user.

However, as in the prior art (S900), in order to photograph a user's living body (iris) using a camera of a mobile device, the user moves the camera or touches his or her face so that his or her iris is placed in a preset recognition area of the camera. Since it has to be moved, the process is cumbersome and complicated, which has the disadvantage of inferior in ease of use.

In addition, looking at the authentication process of step 940 (S940) of the prior art (S900), step 940 (S940) is a method in which, when biometric information is obtained through a camera, the biometric information of registered registrants is individually compared and compared. Authentication is being performed.

However, as in step 940 (S940) of the prior art (S900), the method of simply collating the acquired biometric information with the previously registered biometric information in a fixed order or random order is As the number increases, the authentication elapsed time (Δt), which is the time required for authentication, increases, resulting in a problem that not only decreases the speed of work, but also increases unnecessary time consumption.

The present invention is to solve this problem, and the problem of the present invention is that the controller extracts the iris region corresponding to the distance Δd to the user detected by the distance sensor, and then extracts the iris image from the extracted iris region. By being configured to extract, it is to increase the iris recognition rate and obtain the iris image accurately and quickly, thereby increasing the overall authentication speed, and to provide an HMI device that can reduce unnecessary time consumption for authentication.

In addition, another problem of the present invention is to provide an HMI device that facilitates the user's iris recognition process and enhances the convenience of use because the controller can recognize the iris even if the user does not move the iris to a position close to the camera. will be.

In addition, another problem of the present invention is to apply the iris authentication system to the HMI and perform user authentication by comparing and contrasting the feature pattern of the user's iris image with the feature patterns of the registered users when the user approaches. It is to provide an HMI device that can significantly increase security because only users can operate and access it.

In addition, another problem of the present invention is that the sorting unit for each inquiry order of the controller optimizes the inquiry order of registered feature data at each preset period (T), thereby maximizing the authentication speed compared to the same configuration, as well as unnecessary calculation processing and time. It is to provide an HMI device that can effectively prevent consumption.

In addition, another problem of the present invention is to provide an HMI device capable of optimizing authentication speed by flexibly responding to environmental variables by optimizing the inquiry order information according to various environmental variables at each preset period (T) by the inquiry order setting unit of the controller. To provide.

The solution means of the present invention for solving the above problem is a human machine interface (HMI) including a photographing means for photographing a user to obtain an image, and a controller for performing user authentication by analyzing the image acquired by the photographing means. The controller includes: a database unit configured to preset and store a pixel position detection algorithm for outputting a pixel value that can be determined as an iris region among the images when distance (d) information and an image are input as input values; After analyzing the image and distance (d) input from the photographing means and the distance sensor using the pixel position detection algorithm, extracting the iris region image from the input image, and then extracting the iris image by analyzing the extracted iris region image An image analysis unit, wherein the database unit stores registered feature data of pre-registered users, and the controller detects the feature data by analyzing the iris image input from the image analysis unit, and then stores the detected feature data. A user authentication unit that compares registered feature data stored in the database unit and determines whether a corresponding user is an authorized user or not; When user authentication is successful by the user authentication unit, a control unit that allows access to the user is included, and the registered feature data is a feature pattern of an iris image of a previously registered user, and the user authentication unit detects a preset feature pattern. A feature pattern detection unit for detecting a feature pattern by analyzing the iris image acquired by the photographing means using an algorithm; An authentication processing unit for performing authentication by comparing the characteristic pattern detected by the characteristic pattern detection unit with the registered characteristic data, respectively; When authentication is processed by the authentication processing unit, authentication processing information is generated that includes authentication-processed user information, authentication elapsed time (△t), which is the time required for authentication, and contents for each of preset category lists. Then, an authentication processing information generation unit for storing the generated authentication processing information in the database unit; Including a sorting unit for each inquiry order, the sorting unit for each search order, by referring to authentication processing information stored in the database unit, detects the number of authentication times (N) by category list for each pre-registered user, and then, user information and authentication of each user. An authentication cumulative information generation module that generates authentication cumulative information including an elapsed time (Δt) and the number of authentications per detected category list (N); After applying the current time information to each of the preset category lists, the current time base is set to assign a weight in proportion to the ranking of the category lists of the preset search order information, which is the information of the priority given to each of the preset category lists. Weight setting module; In the authentication accumulated information, when the content of the category list to which the current time information is applied has the same content, the weight set to be assigned to the corresponding category list is multiplied by the number of authentication times (N) of the corresponding category list, and then the multiplication values for each user A summing module for calculating a cumulative value by summing; And a data sorting module that sorts the registered feature data according to the size of the cumulative value calculated by the summing module and then stores the registered feature data in the database unit.

In addition, in the present invention, the photographing means is preferably provided with a high magnification lens.

delete

delete

delete

)을 산출하는 인증경과시간 평균값 산출모듈; 상기 인증경과시간 평균값 산출모듈에 의해 산출된 평균값(

)을 기 설정된 설정값(TH, Threshold)과 비교하여 평균값(

)이 설정값(TH) 이상이면, 조회순서정보의 재설정이 필요하지 않다고 판단하되, 평균값(

)이 설정값(TH) 미만이면, 조회순서정보를 재설정하여야한다고 판단하는 재설정여부 판단모듈을 포함하고, 상기 조회순서 설정부는 상기 재설정여부 판단모듈에 의해 조회순서 재설정이 필요하다고 판단될 때, 조회순서정보를 구성하는 카테고리 목록들 각각의 순서를 변경시켜 조회순서정보를 재구성하는 것이 바람직하다.In addition, in the present invention, the user authentication unit further includes an inquiry order setting unit that is executed every preset period (T), and the inquiry order setting unit is among the authentication processing information stored in the database unit during a period from the present to the previous period (T-1). A previous period information extraction module for extracting the generated authentication processing information; The average value of the authentication elapsed times (Δt) of the authentication processing information during the previous period extracted by the previous period information extraction module (

A) authentication elapsed time average value calculation module that calculates; The average value calculated by the authentication elapsed time average value calculation module (

) Is compared with the preset setting value (TH, Threshold) and the average value (

) Is greater than or equal to the set value (TH), it is determined that resetting of the inquiry order information is not necessary, but the average value (

) Is less than the set value (TH), includes a reset determination module that determines that the inquiry sequence information should be reset, and the inquiry sequence setting unit searches when it is determined that the inquiry sequence reset is necessary by the reset determination module. It is preferable to reorganize the inquiry order information by changing the order of each of the category lists constituting the order information.

)을 산출하는 샘플순서별 평균값 산출모듈을 포함하고, 상기 조회순서정보 생성모듈은 상기 샘플순서별 평균값 산출모듀렝 의해 산출된 샘플순서별 평균값(

)들을 비교하여 가장 높은 평균값(

)을 갖는 샘플순서를 조회순서정보로 재설정하는 것이 바람직하다.In addition, in the present invention, the inquiry order setting unit further includes a simulation module that is executed when it is determined that the inquiry order resetting is required by the resetting determination module, and an inquiry order information generation module, wherein the simulation module is stored in the database unit. A sample pattern extraction module for extracting n sample patterns from among the authentication-processed feature patterns during the period T-1; A sample order combination module for detecting sample information by combining the order of the category lists constituting the inquiry order information by the number of all cases; After substituting the current time information into the category list of the sample orders combined by the sample order combination module, a weight is assigned according to the ranking to the category list of each sample information, and the current time information is applied from the authentication accumulated information. When the content of the category list has the same content, the weight assigned to the category list is multiplied by the number of authentications (N) of the category list, and then the multiplication values for each user are summed to calculate the cumulative value for each user. A simulation module for each sample order, which arranges registration feature data according to the size of the accumulated value, and performs authentication of the sample patterns extracted by the sample pattern extraction module using the registration feature data arranged in each sample order; The authentication elapsed time (△t), which is the time required to authenticate each sample pattern by using the registered feature data arranged by the sample order simulation module, is calculated, and the authentication elapsed time of the sample patterns based on the sample order (△ t)'s mean value (

) To calculate an average value for each sample order, and the inquiry order information generation module includes an average value for each sample order calculated by the average value calculation module for each sample order (

) Compared to the highest average value (

It is desirable to reset the sample order with) to the inquiry order information.

According to the present invention having the above problems and solutions, the controller is configured to extract the iris region corresponding to the distance to the user (Δd) detected by the distance sensor, and then extract the iris image from the extracted iris region. At the same time, the iris image can be acquired accurately and quickly, increasing the overall authentication speed, and reducing unnecessary time consumption for authentication.

In addition, according to the present invention, since the controller can recognize the iris without the user moving the iris to a position close to the camera, the user's iris recognition process can be facilitated, thereby enhancing the convenience of use.

In addition, according to the present invention, when a user approaches by applying the iris authentication system to the HMI, the feature pattern of the iris image of the user is compared and contrasted with the feature patterns of the registered users to perform user authentication, so that only authorized users can operate it. And access is possible, so that security can be remarkably improved.

In addition, according to the present invention, by optimizing the inquiry order of the registered feature data registered in each preset period (T), the sorting unit according to the inquiry order of the controller can maximize the authentication speed compared to the same configuration, as well as effectively reduce unnecessary calculation processing and time consumption. Can be prevented.

In addition, according to the present invention, the inquiry sequence setting unit of the controller optimizes the inquiry sequence information according to various environmental variables for each preset period (T), thereby optimizing the authentication speed by flexibly responding to the environmental variables.

1 is a flow chart showing the HMI program control method disclosed in Korean Patent Publication No. 10-2018-0107548 (name of the invention: HMI program control method through a mobile device).
2 is a block diagram showing an HMI system according to an embodiment of the present invention.
3 is a block diagram showing the HMI of FIG. 2.
4 is a block diagram showing the photographing means of FIG. 3.
5 is an exemplary view for explaining the photographing means of FIG. 4.
6 is a block diagram showing the controller of FIG. 3.
7 is a block diagram illustrating an image analysis unit of FIG. 6.
8 is a block diagram illustrating a user authentication unit of FIG. 6.
9 is a block diagram illustrating a sorting unit according to inquiry order of FIG. 6.
10 is a block diagram showing the accumulated value calculation module for each user of FIG. 9.
11 is a block diagram illustrating an inquiry order setting unit of FIG. 6.
12 is a block diagram showing the simulation module of FIG. 11.

Hereinafter, an embodiment of the present invention will be described with reference to the accompanying drawings.

2 is a block diagram showing an HMI system according to an embodiment of the present invention.

The HMI system 1, which is an embodiment of the present invention, improves security by 1) determining whether the user is authorized and accessed by the HMI 3 through user authentication using the iris, and 2) the acquired user's iris characteristic pattern. When comparing and contrasting with previously registered registration data, it is to maximize the authentication speed by optimally determining the inquiry order of registered data according to the current state.

In addition, as shown in Fig. 2, the HMI system 1 of the present invention is installed according to the process line of the industrial site, performs a specific process according to a pre-designed one, collects raw data and stores it in a memory. Machine tools (5-1), ..., (5-N) and machine tools (5-1), ..., (5-N) are managed and controlled, and at the same time, specified according to the user's request. HMI (Human Machine Interface) (3), which requests operation commands from the machine tool and receives and displays response data corresponding to the requested data from the machine tool (3), HMI (3) and machine tool (5-1), .. ., consisting of a communication network 10 that provides a data movement path between (5-N).

At this time, in FIG. 2, for convenience of explanation, it has been described as an example that the HMI 3 performs data communication directly with the machine tools (5-1), ..., (5-N). A commonly used PLC is connected, and the HMI 3 is configured to transmit and receive data through the PLC of each machine tool.

The communication network 10 provides a data movement path between HMI (3) and machine tools (5-1), ..., (5-N), and in detail, wired cable, RS-232C, RS- Various known wireless communication networks such as 422, RS-485, UCP/IP, TCP/IP, DDE, and OLE can be applied.

3 is a block diagram showing the HMI of FIG. 2.

The HMI 3, as shown in FIG. 3, receives a command from the user as well as a touch screen panel 303 that displays response data according to the user's request, and a facial image (hereinafter, referred to as It consists of a photographing means 305 for acquiring (referred to as a high magnification image), and a controller 301 for managing and controlling these 303 and 305.

At this time, in the present invention, for convenience of explanation, it has been described as an example that the HMI 3 consists of a controller 301, a touch screen panel 303, and a photographing means 305, but the HMI 3 provides lighting when photographing. It may further include a lighting means (not shown) and a monitor (not shown) for separately displaying the high-magnification image obtained by the photographing means 305.

The photographing means 305 is configured to perform narrow-angle photographing and photographs a user's face area to obtain a high magnification image.

FIG. 4 is a configuration diagram showing the photographing means of FIG. 3, and FIG. 5 is an exemplary view for explaining the photographing means of FIG. 4.

As shown in FIG. 4, the photographing means 305 is an iris recognition camera 3051 and a high magnification lens 3052 installed in front of the iris recognition camera 3051 to capture a user's face to obtain a high magnification image. And, it consists of a distance sensor 3053 that measures the distance to the user, the subject.

That is, the photographing means 305 of the present invention is a state where the user is spaced a predetermined distance from the iris recognition camera 3051 as shown in FIG. 5 as the high magnification lens 3052 is installed in front of the iris recognition camera 3051 Even if it is photographed at, the iris recognition camera 3051 not only can acquire a high magnification image of the user's face, but also the area of the iris recognition area S increases in the height direction as the distance to the user increases. As a result, the user does not have to move the iris to a position close to the iris recognition camera 3051, and the iris recognition area S of the conventional camera is narrow so that the user bends the waist or moves the iris. Unnecessary pre-operations such as, etc. are omitted, so that the convenience of use can be significantly improved.

The distance sensor 3053 is a sensing means for measuring a distance to a subject, such as an infrared sensor, and detects the distance Δd to a user.

The photographing means 303 configured as described above outputs the iris region image P acquired by the iris recognition camera 3051 and the distance (Δd) information detected by the distance sensor 3053 to the controller 301 do.

The controller 301 is a computer that manages and controls the overall operation of the HMI 3, requests data from the machine tool 5 at the request of a user, and receives response data corresponding thereto. The received response data is processed and displayed through the touch screen panel 303 equipped with the processed data.

In addition, the controller 301 stores registration data, which is characteristic pattern information of users previously registered in the database unit 31.

In addition, when the controller 301 receives a control request from the user, the controller 301 controls the photographing means 305 to capture the user's face to obtain a high magnification image, and then the iris region from the high magnification image corresponding to the distance Δd from the user. After the image is extracted, the extracted iris region image is analyzed to detect an iris image from the iris region image, the detected iris image is analyzed to detect a feature pattern, and the detected feature pattern is stored in the database unit 31 as registration data. User authentication is performed by comparing with the user, and if authentication fails, the user is judged as an unauthorized person and access is denied.

In addition, the controller 301 optimizes the order of inquiry (contrast) of registration data used for authentication every preset period (T), thereby effectively reducing the authentication elapsed time (Δt), which is the time required for authentication. At this time, the process and method of optimizing the inquiry (contrast) order by the controller 301 will be described in detail with reference to FIGS. 8 to 12 to be described later.

At this time, in the present invention, for convenience of explanation, the contents of the typical service of the HMI 3 are already known technology, and thus will be omitted. For example, the HMI 3 stores graphics designed for each machine tool's appearance and process line so that monitoring data transmission and response data display can be easily performed according to the facility site and each machine tool. At this time, the graphic is directly designed and produced by the user through a graphic design software for PC equipped with a design tool, and the produced design is downloaded to the HMI(3), and the HMI(3) transfers the downloaded graphic to the machine tool(5-1). ), ..., (5-N) are sorted in the same order as the process line order, and then saved, so that users can manage data efficiently through graphics.

6 is a block diagram showing the controller of FIG. 3.

As shown in FIG. 6, the controller 301 includes a control unit 30, a database unit 31, a communication interface unit 39, a user authentication unit 33, a data input unit 34, and a request data generation unit ( 35), the display unit 36, it consists of an additional service unit (37).

The control unit 30 is an OS (Operating System) of the controller 301, and manages and controls the control targets 31, 32, 33, 34, 35, 36, and 37. do.

In addition, when the control unit 30 receives an initial operation request from the user, the image analysis unit 32 is executed, and when an iris image is detected by the image analysis unit 32, the detected iris image is input to the user authentication unit 33. do.

In addition, the control unit 30 1) allows the user to access if the user authentication is successful by the user authentication unit 33, but 2) if the user authentication fails by the user authentication unit 33, the user's access Do not allow. At this time, allowing the user's access means that the HMI(3) activates the interface with the user, and not allowing the user's access means that the HMI(3) deactivates the interface with the user.

In addition, the control unit 30 inputs the data input through the data input unit 34 from the user-authenticated user to the request data generation unit 35, and When the request data is generated, the communication interface unit 39 is controlled so that the generated request data is transmitted to the corresponding machine tool 5.

In addition, when the control unit 30 receives response data corresponding to the request data from the machine tool 5 through the communication interface unit 5, the received response data is input to the display unit 36.

In the database unit 31, previously produced graphic user interfaces (GUIs) are stored.

In addition, in the database unit 31, communication identification information of each of the assigned machine tools 5 is stored as an existing audio source.

In addition, a feature pattern detection algorithm for detecting a feature pattern by analyzing an iris image is preset and stored in the database unit 31.

In addition, the database unit 31 stores registration feature data of each of the previously registered users. At this time, the registered feature data consists of user information, an iris image, and a feature pattern. At this time, in the present invention, it has been described for example that user authentication is performed by detecting a feature pattern of an iris image and comparing it with a previously registered user's feature pattern, but the configuration of the registered feature data is not limited thereto, and It is natural that various known configurations capable of determining the degree of correspondence can be applied.

In addition, the database unit 31 stores authentication processing information generated by the authentication processing information generation unit 334 of FIG. 8 to be described later. At this time, the authentication processing information is composed of authentication-processed user information, authentication elapsed time (Δt), which is the time required for authentication, and contents for each preset category list.

In this case, the category list is made up of types that can classify the authentication point, and in detail, it may be composed of'time zone (time range)','day of the week', and'date'.

7 is a block diagram illustrating an image analysis unit of FIG. 6.

As shown in FIG. 7, the image analysis unit 32 includes a photographing means control module 321, a data input module 322, an iris region image extraction module 323, and an iris image extraction module 324.

The photographing means control module 321 is executed under the control of the control unit 30 when a user requests an initial operation.

In addition, the photographing means control module 321 outputs a trigger signal for operating the photographing means 305 to the photographing means 305. At this time, the photographing means 305, when receiving a trigger signal from the photographing means control module 321, drives the iris recognition camera 3052 to photograph the user's face to obtain a high magnification image.

The data input module 322 receives a high magnification image obtained by photographing from the photographing means 305.

In addition, the data input module 322 receives distance (d) information detected by the distance sensor 3053 from the distance sensor 3053.

The iris region image extraction module 323 detects pixel values corresponding to the distance d input through the data input module 322 using a preset pixel position detection algorithm. In this case, the pixel position detection algorithm is an algorithm that outputs a pixel value that can be determined as an iris region of the image when distance (d) information and an image are input as input values.

In addition, the iris region image extraction module 323 extracts the iris region image by extracting detected pixel values from the high magnification image.

That is, the iris region image extraction module 323 can immediately extract the iris region image including the iris in consideration of the distance to the user from the high magnification image, which is a facial image.

The iris image extraction module 324 extracts an iris image from the iris region image extracted by the iris region image extraction module 323. At this time, since the technology and method for extracting the iris image from the input image is a technology and method commonly used in the iris recognition system, a detailed description will be omitted.

At this time, the iris image extracted by the iris image extraction module 324 is input to the user authentication unit 33 under the control of the controller 30.

When the user authentication unit 33 receives the iris image from the image analysis unit 32, it determines whether the user is an authorized user or not, and if the user is an authorized user, it is determined to allow the access of the user. And, if the user is not an authorized user, it is determined to disallow access to the user.

The operation of the user authentication unit 33 will be described in detail in FIGS. 8 to 12 to be described later.

The data input unit 34 receives a command from a user through the touch screen panel 303. At this time, the command data input through the data input unit 34 is input to the request data generation unit 35 under the control of the control unit 30.

The request data generation unit 35 generates request data corresponding to the command data input through the data input unit 34. At this time, the request data generated by the request data generation unit 35 is transmitted to the machine tool 5 through the communication interface unit 39 under the control of the control unit 30.

The display unit 36 processes the response data transmitted from the machine tool 5 and displays it through the touch screen panel 303.

The additional service unit 37 provides services normally provided by the HMI.

At this time, in FIG. 6, for convenience of explanation, the controller 301 performs the functions of these components including a data input unit 34, a request data generation unit 35, a display unit 36, and an additional service unit 37. Although it has been described as an example, the data input unit 34, the request data generation unit 35, the display unit 36, and the additional service unit 37 use functions commonly used in the HMI 3, for example. It is shown.

8 is a block diagram illustrating a user authentication unit of FIG. 6.

As shown in FIG. 8, the user authentication unit 33 is a characteristic pattern detection unit 331, an authentication processing unit 333, an authentication processing information generation unit 334, an order sorting unit 335 for each inquiry order, and an inquiry order information setting unit. It consists of (336), the register (337).

The feature pattern detection unit 331 analyzes the input iris image using a preset feature pattern detection algorithm and detects a feature pattern of the corresponding iris image. In this case, since the technology and method for detecting the feature pattern by analyzing the subject image is a technology and method commonly used in the image analysis system, a detailed description will be omitted.

The authentication processing unit 333 sequentially compares and contrasts the feature pattern detected by the feature pattern detection unit 331 with registered feature data stored in the database unit 31 to determine whether the user is an authorized user or not. At this time, the registration feature data stored in the database unit 31 are sorted by the sorting unit 335 by inquiry order every preset period (T), and the sorting unit 335 by inquiry order is a query order setting unit ( Registered feature data stored in the database unit 31 are sorted according to the inquiry order information set by 336.

That is, the registration feature data stored in the database unit 31 are sorted and stored in a different search order by the sorting unit 335 for each search order at each preset period (T).

In addition, the authentication processing unit 333, when comparing and comparing each of the detected feature pattern and the registered feature data, 1) when the same registered feature data exists, determines the user of the two data as the same person (determined as an authorized user) It is judged that the authentication has been successful, and 2) when the same registration feature data does not exist, the user is determined to be an unauthorized user, and it is determined that the authentication has failed. At this time, when it is determined by the authentication processing unit 333 that the authentication is successful, the control unit 30 generates the authentication processing information generation unit 334.

The authentication processing information generation unit 334 is executed when authentication is successful by the authentication processing unit 333, and includes authentication processing information including information of a user who has been successfully authenticated, authentication time information, and authentication elapsed time (Δt). Generate.

At this time, the authentication processing information generated by the authentication processing information generation unit 334 is stored in the database unit 31 under the control of the control unit 30.

The registration unit 337 is executed when a request for registration of new registration feature data is received from a user other than a previously registered user, receives registration feature data from the user, and stores the input registration feature data in the database unit 31.

9 is a block diagram illustrating a sorting unit according to inquiry order of FIG. 6.

As shown in FIG. 9, the sorting unit 335 for each inquiry order includes an authentication accumulated information generation module 3351, an accumulated value calculation module 3352 for each user, and a data sorting module 3355.

The accumulative authentication information generation module 3351 generates accumulative authentication information consisting of the number of authentication times (N) for each pre-registered user by using the authentication processing information stored in the database unit 31. In this case, the category list consists of types that can classify the authentication point, and in detail, it may be composed of'time zone (time range)','day of the week', and'date'.

At this time, the authentication processing information is composed of user information, authentication time information, and authentication elapsed time (Δt) information, and the authentication time information includes time zone (time range), day of the week, and date information, which are a list of preset categories.

For example, the accumulative authentication information generation module 3351 uses the authentication processing information collected so far for the user'A', and the user information (A), the time range (13 to 14:00: 5 times, 23 to 24 hours, 12 hours). Times), day of the week (Monday: 9 times, Wednesday: 1 time, Friday: 7 times), date (1 day: 1 time, 2 days: 1 time, 10 days, 5 times, 20 days: 5 times, 30 days: Accumulated authentication information can be created in 5 times).

That is, the accumulative authentication information generation module 3351 generates accumulative authentication information consisting of the number of authentication times (N) for each category of pre-registered users by using the authentication processing information.

10 is a block diagram showing the accumulated value calculation module for each user of FIG. 9.

As shown in FIG. 10, the accumulated value calculation module 3355 for each user includes a first data input module 33531, a second data input module 33532, a current time-based weight setting module 3335, and a summing module. 33534).

The first data input module 33531 receives the authentication accumulated information generated by the authentication accumulated information generating module 3351.

The second data input module 33532 receives the inquiry sequence information set by the inquiry sequence setting unit 336 of FIG. 11 to be described later.

The current time-based weight setting module 33533 applies the current time information to each of the preset category lists. For example, since the current time-based weight setting module 33335 applies the current time to a preset category list, such as time zone (time range), day of the week, and date, the current time of 07 o'clock, Wednesday, and 06 days is used in each category list. Can be applied to.

In addition, the current time-based weight setting module 33335 sets a weight assigned to the category lists according to the inquiry order information input through the second data input module 33532. In this case, a higher score is given to the weight as the search order of the corresponding category list is higher, and the weight for each search order may be preset or may be set according to an input of an administrator.

For example, the current time-based weight setting module 33533 assumes that the inquiry order information is the first priority (time range), the second priority (day of the week), and the third priority (date), the current time-based weight setting module 33335 Gives the highest score, the weight α', to the first priority list (time range), the second highest score, the weight'β' (β<α), to the second priority list (day of the week), and gives the third priority list (date ) Can be set by giving the lowest score, the weight'γ' (γ<β<α).

When the summing module 33532 has the same contents as the contents of each category list to which the current time is applied by the current time-based weight setting module 33335 in the authentication accumulated information input through the first input module 33531, the corresponding The weight set to be assigned to the category list is multiplied by the number of authentication times (N) of the corresponding category list, and then the multiplication values for each user are summed to calculate the cumulative value.

For example, the search order information to which the current status is applied is 1st (13~15:00), 2nd (Saturday), 3rd (1st), and the weights set for each category list are α (time zone), β (day of the week). , γ (date), and the accumulated authentication information of random user A is within the time range (13 to 14:00: 5 times, 23 to 24: 12 times), day of the week (Monday: 9 times, Wednesday: 1 time, Friday: 7 times ), assuming that the date (1 day: 2 times, 2 days: 1 time, 10 days, 5 times, 20 days: 5 times, 30 days: 5 times) Since the number of times user A authenticates is 5 times from 13 to 15:00 when the current time is applied, the number of authentications is multiplied by 5 and the weight'α' assigned to the category list (time zone), and the current time of the second priority category list is applied. Since the number of authentications by User A on Saturday is 0, the number of authentications is 0 and the weight'β' assigned to the category list (day of the week) is multiplied, and User A authenticates on the 1st day when the current time of the 3rd category list is applied Since the number of times is 2, the cumulative value can be calculated by multiplying the number of authentications by 1 and the weight'γ' assigned to the category list (date), and summing the multiplication values '5α', '0', and '2γ'. have.

Returning to FIG. 9 and looking at the data sorting module 3355 of FIG. 9, the data sorting module 3355 sorts the registered feature data according to the size of the accumulated value calculated by the accumulated value calculation module 3355 for each user described above. do.

At this time, the registration feature data sorted by the data sorting module 3355 are stored in the database unit 31.

Returning to FIG. 6 and looking at the inquiry sequence setting unit 336 of FIG. 6, the inquiry sequence setting unit 336 is executed at every preset period T to reset inquiry sequence information.

As shown in FIG. 11, the inquiry order setting unit 336 includes a previous period information extraction module 3331, an authentication elapsed time average value calculation module 3362, a reset determination module 3362, a simulation module 3362, It consists of an inquiry sequence information generation module 3365.

The previous period information extraction module 3331 extracts authentication processing information generated during the previous period (T-1) from the present among the authentication processing information stored in the database unit 31.

)을 산출한다.The authentication elapsed time average value calculation module 3362 refers to the authentication elapsed time (Δt) of the authentication processing information for the previous period input from the previous period information extraction module 3331, and the average value of the authentication elapsed time (Δt) (

) Is calculated.

)을 산출한다.That is, the authentication elapsed time average value calculation module 3362 is an average value of the time required for user authentication made during the previous period (T-1) (

) Is calculated.

)을 기 설정된 설정값(TH, Threshold)과 비교한다. 이때 설정값(TH)은 조회순서정보의 재설정이 필요하다고 판단할 수 있는 인증경과시간의 최소값으로 정의된다.The reset determination module 3362 is an average authentication elapsed time calculated by the authentication elapsed time average value calculation module 3362 (

) Is compared with the preset setting value (TH, Threshold). At this time, the set value TH is defined as the minimum value of the authentication elapsed time that can be determined that resetting of the inquiry order information is necessary.

)이 설정값(TH) 이상이면, 현재의 조회순서정보를 재설정할 필요가 없다고 판단한다. 이때 제어부(30)는 재설정여부 판단모듈(3363)에 의해 조회순서정보의 재설정이 필요하지 않다고 판단하면, 별도의 동작을 수행하지 않아 이전 주기(T-1)에 적용된 조회순서정보가 그대로 적용되도록 한다.The reset determination module 3363 is the average value of the authentication elapsed time (

) Is greater than or equal to the set value TH, it is determined that there is no need to reset the current inquiry order information. At this time, if the control unit 30 determines that resetting of the inquiry order information is not necessary by the reset determination module 3362, a separate operation is not performed so that the inquiry order information applied to the previous period (T-1) is applied as it is. do.

)이 설정값(TH) 미만이면, 조회순서정보를 재설정하여야한다고 판단한다. 이때 제어부(30)는 재설정여부 판단모듈(3363)에서 조회순서정보의 재설정이 필요하다고 판단되는 경우, 시뮬레이션 모듈(3364)을 실행시킨다.In addition, the reset or not determining module 3363 is the average value of the authentication elapsed time (

) Is less than the set value (TH), it is determined that the inquiry order information should be reset. At this time, the controller 30 executes the simulation module 3362 when it is determined that the retrieval sequence information needs to be reset by the reset determination module 3362.

The simulation module 3366 is executed when it is determined that the retrieval sequence information needs to be reset by the reset determination module 3362.

12 is a block diagram showing the simulation module of FIG. 11.

As shown in FIG. 12, the simulation module 3366 includes a sample pattern extraction module 33541, a sample order combination module 33642, a simulation module for each sample order 333643, and an average value calculation module 33644 for each sample order.

The sample pattern extraction module 33641 searches for feature patterns of users who have been authenticated during the previous period T-1 stored in the database unit 31 and randomly extracts n sample patterns among them.

The sample order combination module 33642 detects search order information (hereinafter referred to as sample order) obtained by combining the order of category lists constituting the search order information by the number of all cases.

For example, assuming that the category list is'day of the week','time zone', and'date', the sample order combination module 33642 is a) 1st priority (time range), 2nd priority (day of the week), 3rd priority (date ) And, b) Priority 1 (time range), Priority 2 (date), Priority 3 (day of the week), c) Priority 1 (day of the week), Priority 2 (time range), Priority 3 (date), d)1 Rank (day), rank 2 (date), rank 3 (time range), e) rank 1 (date), rank 2 (time range), rank 3 (day), and rank f) rank 1 (date), 2 Sample orders can be configured with a total of 6 combinations, such as ranking (day of the week) and 3 ranking (time range).

The simulation module for each sample order (33643) is performed by the authentication accumulated information generation module 3351 of the sorting unit 335 for each inquiry order of FIG. 6, the accumulated value calculation module 3355 for each user, and the data alignment module 3355 In the same way, the registration feature data is sorted by sample order.

For example, assuming that the category list is'day of the week','time zone', and'date', the sample order is composed of six combinations, as described above, and accordingly, the authentication module 3763 has a sample order. Since it is composed of 6 combinations, a total of 6 differently sorted registration data (priority applied registration data) are created.

In addition, the simulation module for each sample order (33643) performs authentication by comparing and comparing the sample patterns extracted by the sample pattern extraction module (33641) with registration feature data arranged in the sample order.

In other words, the simulation module for each sample order (33643) performs authentication by using each sample pattern and registration feature data arranged in each sample order.

The average value calculation module 33644 for each sample order calculates the authentication elapsed time, which is the time required to authenticate each sample pattern by using the registration feature data sorted by the sample order simulation module 33643, and samples based on the sample order. The average value of the authentication elapsed times of the patterns is calculated.

For example, when two sample patterns are extracted and the sample order is combined into six, the average value calculation module 33644 for each sample order calculates a total of 12 authentication elapsed times, and a total of six average values are calculated based on the sample order. Can be calculated.

At this time, the average values calculated by the average value calculation module 33644 for each sample order are input to the inquiry order information generation module 3365 of FIG. 8 described above.

Returning to FIG. 11 again, looking at the inquiry sequence information generation module 3365, the inquiry sequence information generation module 3365 compares average values for each sample order input from the simulation module 3362, and the sample sequence having the highest average value among them Is reset to the inquiry order information.

At this time, the inquiry sequence information reset by the inquiry sequence information generation module 3365 is stored in the database unit 31 under the control of the controller 30.

As described above, in the HMI system 1 according to an embodiment of the present invention, the controller 301 extracts the iris region corresponding to the distance Δd to the user detected by the distance sensor, and then extracts the iris image from the extracted iris region. By being configured to extract, it is possible to increase the iris recognition rate and obtain the iris image accurately and quickly, thereby increasing the overall authentication speed, and reducing unnecessary time consumption for authentication.

In addition, since the HMI system 1 of the present invention can recognize the iris of the controller even if the user does not move the iris to a position close to the camera, the user's iris recognition process is facilitated, thereby improving the convenience of use.

In addition, the HMI system 1 of the present invention applies the iris authentication system to the HMI 3 to perform user authentication by comparing and comparing the feature pattern of the iris image of the user with the feature patterns of the registered users when the user approaches. By doing so, only authorized users can operate and access, so that security can be remarkably improved.

In addition, the HMI system 1 of the present invention can maximize the authentication speed compared to the same configuration by optimizing the inquiry order of the registered feature data that the controller 301 of the HMI 3 has previously registered at every preset period (T). In addition, unnecessary computational processing and time consumption can be effectively prevented.

In addition, the HMI system 1 of the present invention allows the controller 301 of the HMI 3 to optimize the authentication speed by flexibly responding to the environmental variables by optimizing the inquiry sequence information according to various environmental variables at each preset period (T). I can.

1: HMI system 3: HMI 5-1, ..., 5-N: machine tools
10: communication network 30: control unit 31: database unit
32: communication interface unit 33: user authentication unit 34: data input unit
35: request data generation unit 36: display unit 37: additional service unit
301: controller 303: photographing means 305: touch screen panel
321: photographing means control module 322: data input module
323: iris area image extraction module 324: iris image extraction module
331: feature pattern detection unit 333: authentication processing unit 334: authentication processing information generation unit
335: sorting unit by inquiry order 336: inquiry order setting unit 337: registration unit
3051: iris recognition camera 3052: high magnification lens 3053: distance sensor
3351: Accumulated authentication information generation module 3352: Accumulated value calculation module for each user
3353: data alignment module 3361: previous cycle information extraction module
3362: authentication elapsed time average value calculation module 3363: reset determination module
3364: reset determination module 3365: simulation module
3366: Inquiry order information generation module

Claims (7)

HMI (Human machine interface):
The controller is
A database unit in which, when distance (d) information and an image are input as input values, a pixel position detection algorithm for outputting a pixel value that can be determined as an iris region of the image is preset and stored;
After analyzing the image and distance (d) input from the photographing means and the distance sensor using the pixel position detection algorithm, extracting the iris region image from the input image, and then extracting the iris image by analyzing the extracted iris region image Including an image analysis unit,
The database unit stores registration feature data of pre-registered users,
The controller is
A user authentication unit that analyzes the iris image input from the image analysis unit to detect feature data, and then compares the detected feature data with registered feature data stored in the database to determine whether the user is an authorized user or not ;
If the user authentication is successful by the user authentication unit, including a control unit that allows the access of the user,
The registered feature data is a feature pattern of an iris image of a previously registered user,
The user authentication unit
A feature pattern detection unit for detecting a feature pattern by analyzing the iris image acquired by the photographing means using a preset feature pattern detection algorithm;
An authentication processing unit for performing authentication by comparing the characteristic pattern detected by the characteristic pattern detection unit with the registered characteristic data, respectively;
When authentication is processed by the authentication processing unit, authentication processing information is generated that includes authentication-processed user information, authentication elapsed time (△t), which is the time required for authentication, and contents for each of preset category lists. Then, an authentication processing information generation unit for storing the generated authentication processing information in the database unit;
Including a sorting unit by order of inquiry,
The sorting unit according to the inquiry order
After detecting the number of authentication times (N) for each category list for each pre-registered user by referring to the authentication processing information stored in the database unit, the user information of each user, authentication elapsed time (△t), and each detected category list An authentication cumulative information generation module for generating authentication cumulative information including authentication counts (N);
After applying the current time information to each of the preset category lists, the current time base is set to assign a weight in proportion to the ranking of the category lists of the preset search order information, which is the information of the priority given to each of the preset category lists. Weight setting module;
In the authentication accumulated information, when the content of the category list to which the current time information is applied has the same content, the weight set to be assigned to the corresponding category list is multiplied by the number of authentication times (N) of the corresponding category list, and then the multiplication values for each user are calculated. A summing module for calculating a cumulative value by summing;
And a data sorting module that sorts the registered feature data according to the size of the accumulated value calculated by the summing module and then stores the registered feature data in the database unit. The HMI according to claim 1, wherein the photographing means comprises a high magnification lens. delete delete delete In claim 1 or 2, the user authentication unit further comprises a query order setting unit that is executed every preset period (T),
The inquiry order setting unit
A previous period information extraction module for extracting authentication processing information generated during a previous period (T-1) from the present among the authentication processing information stored in the database unit;
The average value of the authentication elapsed times (Δt) of the authentication processing information during the previous period extracted by the previous period information extraction module (

A) authentication elapsed time average value calculation module that calculates;
The average value calculated by the authentication elapsed time average value calculation module (

) Is compared with the preset setting value (TH, Threshold) and the average value (

) Is greater than or equal to the set value (TH), it is determined that resetting of the inquiry order information is not necessary, but the average value (

) Is less than the set value (TH), including a reset determination module that determines that the inquiry sequence information should be reset,
When it is determined by the reconfiguration determination module that the retrieval order setting unit requires resetting the retrieval order, reorganizes the inquiry order information by changing the order of each of the category lists constituting the retrieval order information. The method of claim 6, wherein the inquiry order setting unit further comprises a simulation module that is executed when it is determined that the inquiry order resetting is necessary by the resetting determination module, and an inquiry order information generation module,
The simulation module
A sample pattern extraction module for extracting n sample patterns from among the authentication-processed feature patterns during the previous period (T-1) stored in the database unit;
A sample order combination module for detecting sample information by combining the order of the category lists constituting the inquiry order information by the number of all cases;
After substituting the current time information into the category list of the sample orders combined by the sample order combination module, a weight is assigned according to the ranking to the category list of each sample information, and the current time information is applied from the authentication accumulated information. When the content of the category list has the same content, the weight assigned to the category list is multiplied by the number of authentications (N) of the category list, and then the multiplication values for each user are summed to calculate the cumulative value for each user. A simulation module for each sample order, which arranges registration feature data according to the size of the accumulated value, and performs authentication of the sample patterns extracted by the sample pattern extraction module using the registration feature data arranged in each sample order;
The authentication elapsed time (△t), which is the time required to authenticate each sample pattern by using the registered feature data arranged by the sample order simulation module, is calculated, and the authentication elapsed time of the sample patterns based on the sample order (△ t)'s mean value (

), and an average value calculation module for each sample order,
The inquiry sequence information generation module
The average value for each sample order calculated by the average value calculation mode for each sample order (

) Compared to the highest average value (

HMI, characterized in that the sample order with) is reset to the inquiry order information.

Images