KR20130047481A - Driver recognition system and the method for controlling the same - Google Patents

Abstract

PURPOSE: A driver recognition device and a control method thereof are provided to recognize the face of a driver based on the distance, the color, and the section feature of the pupils, thereby storing individual driving environment of the driver by using a single key. CONSTITUTION: A sensing unit(1) includes an image sensor which senses a face image of a driver. A control unit(20) receives signals from the sensing unit and an input unit(5). A storage unit(30) is connected to the control unit and stores driver features. The driver features include a driver pupil feature, a driver pupil color feature, and a driver section feature which show a size of an area with pupils and lips. An operation unit(40) is connected to the control unit and processes the face image of the driver corresponding to an operation control signal of the control unit. [Reference numerals] (1) Sensing unit; (20) Control unit; (3) Wireless communication unit; (30) Storage unit; (40) Operation unit; (5) Input unit; (51) Warning unit; (53) Vehicle startup unit; (55) Mirror motor unit; (57) Seat motor unit; (5a) Mirror switch; (5b) Seat switch; (5c) Mirror connected switch; (5d) Memory switch

Description

Driver recognition device and its control method {DRIVER RECOGNITION SYSTEM AND THE METHOD FOR CONTROLLING THE SAME}

The present invention relates to a cognitive device, and more particularly, to an apparatus and a control method for controlling an operating unit of a vehicle by recognizing a face of a driver of a vehicle.

BACKGROUND ART A vehicle such as an automobile has a function as a variety of convenience means for enabling a user to provide a more stable and comfortable running state beyond a function as a moving means. Accordingly, the vehicle is provided with various convenience facilities, various switches for operating and controlling the same, and display devices thereof.

Various systems are provided for detecting driver's drowsiness due to the convenience of the vehicle or the advanced safety device.

In addition, active research and production of the personal authentication method through the smart key of the vehicle has been made for the driver's security.

However, a smart key for a vehicle has been accompanied with a problem that each passenger must have a different key when making the individual recognition. For example, an advantage of the smart key is that when the holder of the smart key rides through the driver's individual recognition, a stable driving environment is provided by adjusting the position of the seat and the side mirror. However, in the case of a vehicle shared by the family, there is a problem that it is difficult to automatically set the seat or side mirror of the vehicle for each driver using the vehicle smart key.

On the other hand, if the position of the head changes while the driver is driving, the perception area through the vehicle side mirror is changed, causing a possibility of a safety accident.

Accordingly, the present invention provides convenience by making the personal authentication of the driver through the components that make up the image sensing, and when the position of the driver's head, that is, the face moves, detects the positional change through the face recognition to link the side mirrors. An object of the present invention is to provide a driver recognition apparatus and a control method.

According to an aspect of the present invention, there is provided a driver recognition apparatus mounted on a vehicle, including: a sensing unit including an image sensor configured to sense an image of a driver's face; An input unit operable by a passenger; A control unit receiving a signal from the detection unit and the input unit; A driver pupil feature indicating a driver pupil distance, a driver pupil color feature indicating a driver pupil color, and a driver zone feature indicating an area of an area formed by the driver pupil and the lips for determining recognition of a driver's face. A storage unit for storing driver characteristics; And a computing unit connected to the controller and configured to process the driver's face image according to the operation control signal of the controller, wherein the controller compares the driver's face image from the image sensor and the driver's feature to determine the suitability of the driver. It provides a driver recognition device characterized in that the authentication, and controls whether or not the operation unit provided in the vehicle.

In the driver recognizing apparatus, the operation unit may include a warning unit configured to execute a warning by applying a warning control signal from the controller when the controller is not certified for driver's suitability, and according to whether or not the controller conforms to driver's suitability certification. A vehicle starting unit receiving a vehicle start control signal from a control unit, a mirror motor unit receiving a vehicle side mirror control signal from the control unit according to whether or not the controller conforms to a driver suitability certification, and the control unit according to whether or not the controller conforms to a driver suitability certification It may also include a seat motor unit for receiving a vehicle seat control signal from the.

In the driver recognition apparatus, the input unit may include a mirror switch for adjusting the direction of the side mirror of the vehicle, a seat switch for adjusting the seat position of the vehicle, and mirror data preset in the storage unit. Mirror interlock switch to set the mirror interlock mode to adjust the direction of the side mirror of the vehicle according to the movement of the face position, the position of the vehicle side mirror and the driver seat adjusted by the mirror switch and the seat switch input by the driver It may also include a memory switch for storing in the storage unit.

According to another aspect of the invention, the invention is mounted to the vehicle, the sensing unit including an image sensor for detecting a driver's face image; An input unit operable by a passenger; A control unit receiving a signal from the detection unit and the input unit; A driver pupil feature indicating a driver pupil distance, a driver pupil color feature indicating a driver pupil color, and a driver zone feature indicating an area of an area formed by the driver pupil and the lips for determining recognition of a driver's face. A storage unit for storing driver characteristics; And a computing unit connected to the controller and configured to process the driver's face image according to the operation control signal of the controller, wherein the controller compares the driver's face image from the image sensor and the driver's feature to determine the suitability of the driver. Providing a driver recognition device for authenticating and controlling whether an operation unit provided in the vehicle is operated; and the wireless communication unit of the vehicle receives key authentication information from a smart key possessed by a driver and transmits key authentication information. A key authentication step of transmitting to a control unit, the control unit performing driver key authentication by comparing the preset key information stored in the storage unit, and a driver feature for face recognition of the driver using the image information detected by the detection unit; And a setting step of storing the image information detected from the detection unit. Comparing the driver's characteristics that are stored in the group storage unit, and provides the operator recognition device control method comprising a mode execution step of executing an operation mode according to the comparison result.

In the driver recognition device control method, the driver recognition device further comprises a warning unit for executing a warning by applying a warning control signal from the control unit, if the control unit is not certified for the driver, the key authentication step The wireless key information communication step of receiving, by the wireless communication unit, the key authentication information transmitted from the smart key, and the control unit applies the key information control signal to the operation unit to receive the predetermined key information. A key authentication warning step of executing a warning by applying a warning control signal to the warning part when the key authentication information and the preset key information are not identical in the key information checking step and the key information checking step It may also include.

In the driver recognition device control method, the setting step may include: setting an authentication mode for inputting a driver characteristic through the input unit, and setting a forward and backward movement range of a position at which the driver moves on the vehicle seat to set the interlocking range of the vehicle side mirror; A setting mode input step of transmitting a selection mode signal input among setting setting interlocking modes to the control unit; a position adjustment input step of receiving a position signal adjusted by the vehicle seat through a seat switch of the input unit; and the setting mode input A setting mode determination step of determining a setting mode input in the step may be included.

The control method of the driver recognition apparatus may further include executing a setting authentication mode of confirming and storing a driver characteristic from the face image information of the driver when the setting mode determined in the setting mode determination step is a setting authentication mode.

In the controlling method of the driver recognition apparatus, the setting authentication mode executing step may include: a setting authentication detecting step in which the control unit applies a detection control signal to the detection unit to acquire a face image of the driver; A setting image processing step of image-processing the driver's face image detected in the setting authentication detecting step, and a driver's pupil feature indicating a pupil's pupil distance from the driver's face image processed by the operation unit, a color of a driver's pupil, a driver's pupil And a setting authentication feature extraction step of extracting a driver zone feature representing an area of an area formed by the lips, and a setting authentication information storage step of storing the driver feature extracted in the setting authentication feature extraction step in the storage unit. .

In the control method of the driver recognition apparatus, when the setting mode determined in the setting mode determination step is a setting interlocking mode, confirming and setting the interlocking range of the side mirror of the vehicle from the face image information of the driver in the front and rear moving area of the driver The setting interlocking mode may be further included.

In the controlling method of the driver recognition apparatus, the setting interlocking mode execution step may include: a forward movement checking step of acquiring a face movement image in a forward movement area of a driver's face to identify a forward movement area, and the detection unit A face movement image for acquiring a face movement image in the back movement region of the driver's face to confirm the back movement region, and a face movement image for image processing the face movement images acquired in the forward movement confirmation step and the backward movement confirmation step A mirror angle calculating step of calculating mirror angles corresponding to the front and rear moving regions of the driver using the processing step, mirror data preset in the storage unit, and the front and rear of the driver calculated in the mirror angle calculating step Store the mirror interlocking angle state according to the moving area in the storage unit It may comprise a storage step mirror angle interlock state.

In the driver recognition apparatus control method, the forward movement checking step may include: a forward movement instruction step of requesting a forward movement of a driver's face to the driver through the warning unit; The detecting unit may include a forward movement image acquisition step of acquiring a driver forward movement image by sensing an image of the driver's face, and the backward movement checking step may include: requesting the driver to move backward of the driver's face through the warning unit. The rear movement instruction signal may include a rear movement instruction step and a rear movement image acquisition step of acquiring a driver rear movement image by sensing the image of the driver's face when the driver moves backward.

In the method of controlling a driver recognition apparatus, the mode executing step may include: an authentication mode in which the controller compares the driver information stored in the storage unit with the image information detected by the detection unit and an authentication mode for authenticating a driver's face; The operation mode determination step may include determining an operation mode to be executed among operation modes including a mirror interlocking mode for adjusting a position of the vehicle side mirror according to a driver's movement by detecting a face image.

In the driver recognition device control method, when it is determined that the operation mode to be executed in the operation mode determination step is an authentication mode, the authentication mode includes: an authentication detection step of acquiring a face image of the driver and a face image of the driver; The authentication image processing step of the image processing unit, the operation unit, the driver pupil authentication feature indicating the distance of the pupil of the driver from the driver face image processed image, the driver pupil color authentication feature indicating the driver eye color and the driver eyes and lips An authentication feature extraction step of extracting a driver authentication feature including a driver zone authentication feature representing an area of an area to be formed, a driver authentication feature comparison step of comparing the driver authentication feature with a driver feature stored in the storage unit, and the control unit Being said driver And an authentication feature confirming step of confirming the feature compared in the feature feature comparing step, and an authentication data execution step of applying, by the control unit, an operation control signal to the operation unit according to the result of the checking in the authentication feature checking step. have.

In the driver recognition apparatus control method, the extracting of the authentication feature comprises: a driver pupil authentication feature extraction step of extracting a driver's pupil feature from the image of the driver's face processed image and a pupil of the driver from the image of the driver's face image processed; A driver pupil color authentication feature extraction step of extracting color features and a driver zone feature extraction step of extracting a driver zone authentication feature using the driver pupil authentication feature and driver lips may be included.

In the driver recognition apparatus control method, the authentication feature comparison step may include: an authentication pupil interval comparison step of comparing the driver pupil interval by comparing the driver pupil authentication feature and the driver pupil feature stored in the storage unit, and the authentication pupil Authentication pupil color comparing the pupil pupil color by comparing the pupil pupil color identification feature with the driver pupil color authentication feature and the driver pupil color feature stored in the storage unit, to verify whether the comparison result is the same in the interval comparison step. The driver's area area may be compared by comparing the verification pupil color identification step of checking whether the comparison result is the same in the authentication pupil color comparison step, and the driver zone authentication feature and the driver zone feature stored in the storage unit. Comparison of certification area area to compare The comparison result from the system and comparing the authentication zone area may comprise an authentication section area equal to decision step to determine whether they are the same.

In the driver recognition device control method, when it is determined that the operation mode to be executed in the operation mode determination step is an interlocking mode, the interlocking mode includes: an interlocking detection step of acquiring a face image of the driver and a face image of the driver; The interlocking image processing step of the image processing unit, and the operation unit, the driver pupil interlocking feature indicating the distance of the pupil of the driver from the driver face image processed image and the driver zone interlocking feature indicating the area of the area formed by the driver pupil and the lips An interlocking feature extracting step of extracting a driver interlocking feature, a driver interlocking feature comparison step of comparing the driver interlocking feature and the driver feature stored in the storage unit, and a feature that the control unit compares in the driver interlocking feature comparing step Check the interlocking features to confirm And an interlocking data execution step of applying, by the control unit, an operation control signal to the operation unit according to the confirmation result in the interlocking feature checking step.

The driver recognition device control method may include: extracting a driver's pupil feature from the driver's pupil image from the image of the driver's face, and using the driver's pupil interlocking feature and the driver's lips. A driver zone feature extraction step of extracting a driver zone interlocking feature may be included.

In the driver recognition device control method, the interlocking feature comparison step may include: a horizontal size checking step of checking a horizontal size of the driver zone interlocking feature and a horizontal size of the driver zone feature among the authenticated driver features and the driver zone interlocking feature; A horizontal size ratio calculating step of calculating a ratio of a horizontal size of the horizontal size error, a horizontal size error judging step of comparing a horizontal size ratio error range among the preset data pre-set and stored in the storage unit, and the horizontal size ratio; When it is determined that the size error is normal, the vertical size checking step of checking the vertical size of the driver zone interlocking feature and the ratio of the vertical size of the driver zone feature and the vertical size of the driver zone interlocking feature among the certified driver features are determined. A vertical scale ratio calculating step of calculating a value; Pre-set may comprise a vertical size ratio of the error range of the preset data to be stored and the vertical size error determination step of comparing the longitudinal size ratio.

Driver recognition apparatus and control method thereof according to the present invention having the configuration as described above has the following effects.

First, the driver recognition apparatus and its control method according to the present invention, by recognizing the driver's face recognition using the pupil distance, the color of the pupil and the zone feature, it is possible to store the driver's individual driving environment even if using a single key Driver comfort may be provided by automatically adjusting the position of the seat mirror or the side mirror according to the individual driver.

Secondly, the driver recognition apparatus and control method thereof according to the present invention may enhance vehicle security by achieving individual recognition through driver's face recognition.

Third, the driver recognition apparatus and the control method thereof according to the present invention, by recognizing the change of the driver's zone characteristics in conjunction with the movement of the forward and backward position of the authenticated driver to achieve a change in the angle of the corresponding side mirror, etc. It is also possible to ensure a safe driving state by more smoothly.

While the present invention has been particularly shown and described with reference to exemplary embodiments thereof, it will be understood by those of ordinary skill in the art that various changes in form and details may be made therein without departing from the scope of the present invention. Accordingly, the true scope of the present invention should be determined by the technical idea of the appended claims.

1 is a schematic perspective view illustrating an arrangement state of a sensing unit of a driver recognition apparatus according to an embodiment of the present invention.
2 is a schematic block diagram of a driver recognition apparatus according to an embodiment of the present invention.
Figure 3 is a schematic state diagram showing the features of the driver's face utilized by the driver recognition apparatus according to an embodiment of the present invention.
4 to 6 are schematic front and side views showing a face recognition state from a sensing unit of a driver recognition apparatus according to an embodiment of the present invention when the driver moves from the right position to the front and rear positions.
7 to 9 are state diagrams illustrating an example of a change of a zone feature set by a driver recognition apparatus according to an embodiment of the present invention when the driver moves from the correct position to the front and rear positions.
10 to 19 are flowcharts illustrating a method for controlling a driver recognition apparatus according to an embodiment of the present invention.
20 and 21 are state diagrams showing the state of the vehicle side mirror according to the position of the face of the driver of the present invention.

Hereinafter, a driver recognition apparatus according to the present invention will be described with reference to the drawings.

Driver recognition apparatus according to an embodiment of the present invention comprises a detection unit 1, the input unit 5, the control unit 20, the storage unit 30, the operation unit 40, the operation unit 50 Include. In addition, the driver recognition apparatus 10 may further include a wireless communication unit 3 for wireless communication with the vehicle smart key A carried by the driver.

The detector 1 includes an image sensor for acquiring an image of a driver's face, and the detector 1 implemented as an image sensor is mounted in a vehicle interior. As shown in FIG. 1, the sensing unit 1 may be mounted on a dashboard to acquire an image of the driver's face, may be mounted on a vehicle ceiling, may be mounted on a rear view mirror, and the like. Various configurations are possible in the range of acquiring the face image of the camera.

The input unit 5 transmits an input signal input by the occupant, that is, the driver, to the controller 20, and the input unit 5 reflects the driver's intention through the input signal. As shown in FIG. 2A, the input unit 5 includes a mirror switch 5a for adjusting the position of the side mirror of the vehicle, a seat switch 50b for adjusting the position of the seat of the vehicle on which the driver is seated, and The storage unit 30 stores a mirror linkage switch 5c for executing a mirror linkage function in which an adaptive control gear vehicle side mirror is adjusted according to a movement of a face position, and side mirrors or seats according to individual drivers. It may include a memory switch (5d) for storing in. In the present embodiment, the memory switch 5d is used as a selection switch for selecting a setting mode, and the input unit may further include a separate setting mode switch in some cases.

The control unit 20 is connected to the sensing unit 1 and the input unit 5 to transfer a predetermined sensing or input sensing signal and input signal to the operation unit 40 to generate a calculation process or transfer to the storage unit 30. To store predetermined information.

The storage unit 30 may be connected to the control unit 20 and store driver characteristics. The driver characteristics refer to driver information stored for authentication of an individual driver. Driver features include a driver pupil feature, a driver pupil color feature, and a driver zone feature. As shown in Figs. 3 and 4, the driver pupil feature represents the distance Ho of the pupil on the acquired image information of the driver, and the driver eye movement color feature is the color of the driver's eye movement, i.e. black, blue The color information of the pupil, such as a color, is displayed, and the driver zone characteristic represents area information of an area formed by the line Vo formed by the driver's pupil, the height Vo formed by the driver's lips, and the pupil gap Ho of the driver pupil characteristic. In the drawings, Ho is illustrated as a length including the pupil, but in some cases, various modifications are possible, such as taking a reference to measure the horizontal length Ho based on the center of the pupil.

5 and 6 show a front view and a side view showing a state in which the transverse length and the longitudinal length when the front and rear position shifts of the driver's face occur, and the area area thereof are changed. That is, when the driver's face moves forward from the correct position of FIG. 4, as shown in FIGS. 5A and 5B, the distance from the image sensor (driver detection sensor) of the sensing unit 1 is shortened. The size in the image information acquired by the sensing unit 1 becomes larger than at the normal position. That is, the distance between the pupil of the driver has a value of H1 (H1> Ho), and the vertical length between the line segment of the driver's eyes and the horizontal segment of the lips has a value of V1 (V1> Vo). The area characteristic of D1 is larger than the area Do of the area formed by Ho and Vo, and the area for this area. On the other hand, when the driver's face moves backward, as shown in FIG. 6, the distance between the driver's face and the sensing unit 1 is farther away, and thus the values for the horizontal length, the vertical length, and the area area formed by the driver at the home position are shown. It has smaller horizontal length H2, vertical length V2 and zone area D2.

7 to 9, there is a difference in the driver's face, that is, the width of the driver's pupil which is the driver's pupil, which is described below, the length of the horizontal segment formed by the left and right pupils of the driver's zone characteristics, and the horizontal portion of the lips. , And an example of a case where a face of a driver having different areas, which are different from each other, changes in distance from a sensing unit. That is, FIG. 7 illustrates an embodiment of a front-back position change with respect to the driver when the driver is 4cm and 4cm in length, and FIG. 8 is 6cm and 4cm, respectively, and FIG. 9 is 4cm and 6cm. . In each of these cases, image information is acquired for each case of moving backward 10 cm and 20 cm forward based on the position of the normal posture. It was confirmed that the change in the ratio was reached within a certain range. That is, the driver pupil feature and driver zone feature representing the pupil gap according to the front and rear positions of the driver's face, that is, the distance difference from the sensing unit 1 are variable, but the horizontal length (Ho, 1, 2) and the vertical length (Vo) are variable. The ratio of, 1,2 has a constant value and can be used to link the side mirrors by detecting the front and rear position movement state of the driver's face through the ratio.

The calculator 40 is connected to the controller 20 and processes the driver's face image from the image sensor of the detector. In addition, by comparing the driver feature information extracted from the driver face image information processed in accordance with the operation control signal of the control unit 20 and the driver feature stored in the storage unit 30 to authenticate the driver's suitability, Accordingly, the operation of the operation unit 50 mounted on the vehicle is controlled. The operation unit 50 includes a warning unit 51, a vehicle starting unit 53, a mirror motor unit 55, and a seat motor unit 57. The warning unit 51 includes various audiovisual devices such as a display and a speaker. It may be implemented. The vehicle starting unit 53 may be implemented as a device for controlling the on / off of the starting motor or the starting motor for starting the vehicle, and the mirror motor unit 55 is implemented as a motor for operating the position of the side mirror of the vehicle. The seat motor unit 57 is implemented as a motor for adjusting the position of the seat on which the driver is seated. Each of the warning unit 51, the vehicle starting unit 53, the mirror motor unit 55, and the seat motor unit 57 of the operation unit 50 includes a warning control signal from the control unit 20, a vehicle start control signal, and a vehicle. A predetermined operation is made in accordance with the side mirror control signal and the vehicle seat control signal.

In particular, the calculation unit 40 performs an image processing unit 41, a driver information processing unit 42, a key information checking unit 43, a driver drowsiness processing unit 44, a driver data authentication unit 45, and a mirror synchronization processing unit 46. The image processor 41 may process the image information detected by the sensor 1 to convert the image information into binarization data so as to extract driver information for comparison with a driver feature stored in the storage unit 30. .

The driver information processor 42 extracts driver information, that is, driver features, from the image information processed by the image processor 41.

The key information confirming unit 43 receives key authentication information transmitted from the smart key A possessed by the driver according to the key authentication control signal of the control unit 20, and preset key information stored in the storage unit 30 and Compare.

The driver drowsiness processing unit 44 calculates a driver feature extracted from the driver's face image information processed according to the control signal of the drowsiness check control signal of the controller 20, and compares the current drowsiness of the driver. The conventional method may be used, such as a method of checking drowsiness by changing the number of pixels formed by the white of the pupil.

The driver data authentication unit 45 compares the driver feature obtained from the image processing of the driver's face image with the driver feature stored in the storage unit 30, and the mirror synchronization processor 46 is configured to perform an interlocking operation according to the movement of the driver's position. When the mirror interlocking signal is input, the controller 20 transmits the mirror interlocking function to change the position of the side mirror according to the movement of the face position of the driver.

A driver face recognition apparatus control method according to the present invention will be described with reference to the drawings. As shown in FIG. 12, the method for controlling a driver face recognition apparatus includes a provision step S1, a key authentication step S10, a setting step S20, and a mode execution step S30. In some cases, as shown in Figs. 11 and 13, the provisioning step S1, the key authentication step S10 and the setting step S20 are provided or provided (S1), the key authentication step S10 and A configuration including a mode execution step S30 may be taken. This embodiment will be described with reference to FIG. 12.

First, the driver face recognition apparatus 10 according to the present invention is provided in the providing step S1, which is the same as described above, and replaces the above.

Thereafter, the controller 20 executes the key authentication step S10. In the key authentication step S10, the controller 20 determines whether the smart key possessed by the driver is appropriate. That is, in the key authentication step (S10), the wireless communication unit 3 of the vehicle receives the key authentication information transmitted from the smart key (A) possessed by the driver and transmits it to the controller 20. The control unit 20 transmits the received key authentication information to the key information checking unit 43 of the operation unit 40 together with the preset key information stored in the storage unit 30 to confirm whether the key information of the two are the same. . As shown in FIG. 14, the key authentication step S10 includes a wireless key information communication step S11, a key information checking step S13 and S15, and a key information authentication warning step S17. In the wireless key information communication step S11, the control unit 20 applies a reception control signal so that the wireless communication unit A receives the key authentication information transmitted from the smart key A, and the control unit 20 confirms the key information. The key authentication information received at S13 and 15 and the preset key information stored in the storage unit 30 are transmitted to the key information confirming unit 43 to execute a predetermined comparison confirming step. If it is determined in step S13, 15 that the key information does not match, the control unit 20 applies a vehicle warning control signal to the warning unit 51 so that a smart driver that does not have a suitable driver currently or includes suitable key information. An acoustic or visual indication to carry a key may be executed, or in some cases, a function of stopping the starting of the vehicle may be performed by applying an operation stop signal to the vehicle starter 53.

On the other hand, if it is authenticated with the appropriate key information in steps S13 and 15, the control flow goes to step S20.

In the setting step S20, the controller 20 stores the driver's feature for face recognition of the driver using the image information detected by the detector 1. That is, the setting step S20 includes a setting mode input step S21, a position adjustment input step S23, and a setting mode determination step S25. The setting mode input step S21 includes a driver through the input unit 5. The control unit 20 transmits a selection mode signal input between a setting authentication mode for inputting a feature and a setting interlocking mode for setting a linkage range of the vehicle side mirror by setting a forward and backward movement range of a position at which the driver moves on the vehicle seat. . Here, the input unit 5 may include a separate setting switch, but the memory switch 5d of FIG. 2A may be set to perform a function of the setting switch.

The setting step includes two selectable setting modes, which include a setting authentication mode and a setting interlocking mode. The setting authentication mode refers to a mode in which a feature in the image acquired image information is extracted and stored in the storage unit 30 to set and store the driver feature, which is the driver's information, and the setting interlocking mode refers to the driver's information about the driver's information. Detects a fluctuating area as the face moves forward and backward, and when the position of the driver's face changes within the fluctuating area, changes the position of the vehicle side mirror to prevent the side attention from falling due to the fluctuation of the driver's face. Refers to a mode for setting the driver's face moving area as a premise for executing the mirror interlocking mode.

Therefore, the setting mode S20 includes a setting authentication mode execution step S27 and a setting interlocking mode execution step S29, which are each selectively selected according to the determination result in the setting mode determination step S25. In the setting authentication mode execution step (S27), the control unit 20 confirms and stores the driver characteristics from the driver's inferior image information, and in the setting interlocking mode execution step (S29), the control unit 20 controls the driver's motion in the front and rear movement regions of the driver. The interlocking range of the side mirror of the vehicle is confirmed and set from the face image information.

The setting authentication mode execution step S27 includes a setting authentication detection step S271, a setting image processing step S273, a setting authentication feature extraction step S275, and a setting authentication information storing step S277. In the setting authentication detection step S271, the control unit 20 applies a detection control signal to the detection unit 1 so that the detection unit 1 acquires an image of the driver's face, and the setting image processing step S273 includes the control unit 20. ) Transfers the acquired image information to the image processor 41 of the calculator 40 to process the image. Thereafter, the control unit 20 applies a control signal to the driver information processing unit 43 of the operation unit 40 so that the area of the area formed by the driver pupil feature, the driver color feature, the driver pupil, and the lips from the image information of the driver is processed. A setting authentication feature extraction step (S275) of extracting a driver zone feature representative of FIG. Here, the driver pupil feature identifies the pupil of the driver in the processed image information and refers to the pupil interval that is the distance between the left and right pupils. Such pupil spacing may be extracted by extracting pixels of the eye white of the image information, identifying the center of the pupil with the pupil, and calculating the intervals thereof. The driver's pupil color feature calculates an RGB coordinate value by comparing the RGB chromaticity of the image information acquired when the sensor 1, which is an image sensor, acquires color image information to the RGB chromaticity stored in the storage unit 30. It can also be saved as a color feature.

In addition, the driver's zone feature is to secure the position of the lips from the image information, and the rectangular area formed by the vertical length Vo between the line segment formed by the lips and the line segment formed by the pupil and the horizontal length Ho formed by the extracted driver pupil. The driver zone feature can be extracted by setting to the area zone and calculating the area value in this area zone.

The driver feature extracted in the setting authentication feature extraction step S275 is stored as the driver feature in the storage unit 30. That is, the driver's pupil feature, the driver's pupil color feature, and the driver's zone feature are stored in the storage unit 30 together with the position of the vehicle driver's seat and the side mirror's position information input in step S23.

On the other hand, when the control unit 20 determines in step S25 that the setting mode is the setting interlocking mode, not the setting authentication mode, the control unit 20 executes the setting interlocking mode execution step (S29). The setting interlocking mode execution step (S29) includes the forward movement checking step (S291), the rear movement checking step (s293), the face moving image processing step (s295), the mirror angle calculating step (S287), and the mirror interlocking angle state storing step (S299). ).

In the forward movement checking step S291, the sensing unit 1 acquires a face movement image in the front movement area of the driver's face to confirm the front movement area, and in the rear movement checking step S293, the detection unit 1 The face movement image is acquired by acquiring a face movement image in the back movement area of the driver's face, and the face movement image processing step S295 is a face movement acquired in the forward movement confirmation step S291 and the backward movement confirmation step S293. Image processing, and in the mirror angle calculation step (S297) using the mirror data preset in the storage unit 30 to calculate the mirror angle corresponding to the front and rear moving area of the driver, In S299, the control unit 20 stores the mirror 30 in the storage unit 30 in a different mirror interlocking angle state in the front and rear movement regions of the driver calculated in the mirror angle calculation step.

The forward movement confirming step S291 includes a forward movement instructing step S2911 and a forward moving image checking step S2913. The controller 20 controls the driver at the current position through a warning unit or a separate display device or sound output unit. Request movement to the area that can move the most forward of the face. Then, the control unit 20 causes the detection unit 1 to acquire the image information after a predetermined time.

In addition, the backward movement confirming step S293 includes a backward movement instructing step S2931 and a backward movement image confirming step S2933, which means that only the backward movement instead of the forward movement of the driver's face in the forward movement confirming step S291 is performed. It takes the same structure but different, so duplicate description is omitted.

Thereafter, the controller 20 causes the image processor 41 of the calculator to perform image processing on the moving image (S295) and extract a driver feature in the image information through a predetermined process. In this case, when there is no change in the image, the controller 20 may separately display or warn the driver that there is no change in the image. The image processing and feature extraction here are replaced with the above almost same as in the setting step.

Thereafter, the controller 20 calculates mirror angles corresponding to the driver's forward and rearward moving regions by using the mirror data preset in the storage unit 30. Here, the mirror angle may be calculated through an interpolation of the optimized mirror data of the rear view through the side mirror at a predetermined face position of the driver through repeated experiments. That is, as shown in FIG. 21, the angle data of the side mirror for achieving an optimized rearward gaze for the area where the driver's head Dr is movable forward and backward may be stored in the storage unit 30. To the angle face movement distance to change the angle of the side mirror 55a by operating the mirror motor unit 55 using the front and rear movement distances of the face at the current position of the currently occupied driver obtained from the data and the moving image information. Calculate the side mirror angle accordingly. For example, if the maximum moving distance in the moving area from the "back" position to "forward (20 cm)" in FIG. 21 is 40 cm in total and the angle range of the side mirror is from 15 degrees to 50 degrees, Based on the driver zone feature, the vehicle seat position and the side mirror position value for the driver feature set for the driver feature, the proportion of the driver's feature obtained from the current moving image information is compared with the driver zone feature to determine how much the front and rear position movement of the face has occurred. The angle of the side mirror corresponding to the forward and backward position shift values may be derived by using mirror data preset in the storage unit 30. Of course, in this case, it may also be derived in a linear relationship with respect to the maximum travel distance in the moving region from the "back" position to the "forward (20 cm)", and in the angular range at several positions such as "forward (10 cm)". Various modifications are possible, such as being partitioned and derived through interpolation.

In this way, the mirror angle calculated in step S297 is stored in the storage unit 30 together with the corresponding driver characteristics, and the driver later authenticates the driver to quickly move the front and rear sides of the side mirror at a predetermined position during the boarding and driving process. It is also possible to change the position, ie the angle, to make the rear side look of the vehicle more smooth.

The mode execution step S40 includes an operation mode determination step S31 for determining whether the operation mode to be executed is the authentication mode or the mirror interlocking mode. That is, the authentication mode refers to an operation mode in which the control unit 20 authenticates the driver's face by comparing the image information detected by the detection unit 1 with the driver's features stored in the storage unit 30, and the mirror interlocking mode is authentication. It refers to an operation mode that continuously detects the face image of the driver and adjusts the position of the vehicle side mirror according to the driver's movement. If it is determined in the operation mode determination step S41 that the operation mode to be executed is the authentication mode, the control unit 20 executes the authentication mode. The authentication mode includes an authentication detection step S310, an authentication image processing step S311, an authentication feature extraction step S313, an authentication feature comparison step S315, an authentication feature checking step S317, and an authentication data execution step. (S319). In the authentication detection step S311, the control unit 20 applies a detection control signal to the detection unit 1 to obtain image information of the driver's face, and in the authentication image processing step S313, the control unit 20 acquires an authentication feature. The operation control signal is applied to the image processing unit 41 and the driver information processing unit 42 of the calculating unit 40 to extract the image information obtained, and the driver's feature information is obtained therefrom. As illustrated in FIG. 16, the authentication feature extraction step S313 includes a driver pupil authentication feature extraction step S3131, a driver pupil color authentication feature extraction step S3133, and a driver zone authentication feature extraction step S3135. In operation S3131, the driver 20 extracts the driver's pupil feature, that is, the pupil interval, and extracts the driver's pupil color authentication feature in operation S3133, according to the operation control signal of the controller 20. The information processing unit 42 extracts the pupil color information of the driver, and in the driver zone authentication feature extraction step S3135, the horizontal distance Ho formed by the pupil interval and the vertical distance between the line segment formed by the left and right pupils and the line formed by the lips. Extract information on the area area formed by Vo).

Then, the controller 30 compares these driver authentication features with the driver features pre-stored in the storage unit 30 and performs an authentication feature comparison step S315 for comparing whether the corresponding driver is a suitable driver for the storage unit 30. Run The authentication feature comparison step (S315) includes an authentication pupil interval comparison step (S3151), an authentication pupil interval equal determination step (S3152), an authentication pupil color comparison step (S3153), an authentication pupil color identical determination step (S3154), The authentication zone area comparison step S3155 and the authentication zone area equal determination step S3156 are included. In other words,

* The certified pupil interval comparison step (S3151) compares the driver pupil authentication feature and the driver pupil feature of the driver stored in the storage unit 30 to compare the driver pupil interval, that is, the horizontal distance Ho, and the control unit 20 performs the step In S3152, it is determined whether the intervals are the same. If it is determined in step S3152 that the authentication pupil intervals are the same, the control unit 20 proceeds to step S3153. In the authentication pupil color comparison step (S3153), the driver's pupil color authentication feature is compared with the driver's pupil color stored in the storage unit 30, and it is determined in step S3153. Similarly, if the pupil colors match, it is determined whether there is a driver having the same driver zone characteristic as the driver zone authentication feature among the driver features determined to be the same in steps S3152 and S3154. If it is determined in step S3152, S3154 and S3156 that the respective conditions are not the same, the control unit 20 applies the control signal to the warning unit 51 to execute the warning mode S3157. In the present embodiment, each comparison process is based on whether or not it is the same, but whether or not the same may include a case where the comparison includes a range allowing a predetermined error. That is, it is possible to take a structure that judges not only the exact same numerical value but also the substantially identical both when matched within some preset error range.

Then, in step S317, information on the driver having the predetermined authentication characteristic is confirmed, and the control unit 20 confirms the predetermined operation control signal setting to be applied to the operation unit 50 based on this. Thereafter, the control unit 20 applies the operation control signal set in step S317 to the operation unit 50 in step S319 to execute a predetermined authentication data execution mode (S319). Through such operation control, the position of the seat and side mirrors of the vehicle may be moved to a position preset by the recognized driver.

On the other hand, if it is determined in operation S311 that the operation mode is the mirror interlocking mode, the control unit 20 executes the mirror interlocking mode. The interlocking mode includes the interlocking detection step (S320) and the interlocking image processing step (S321), the interlocking feature extraction step (S323), the interlocking feature comparison step (S325), the interlocking feature checking step (S327), and the interlocking data execution step. (S329). In operation S321, the control unit 20 applies a detection control signal to the detection unit 1 to obtain image information of the driver's face, and in operation S323, the control unit 20 provides an interlocking feature. The operation control signal is applied to the image processing unit 41 and the driver information processing unit 42 of the calculating unit 40 to extract the image information obtained, and the driver's feature information is obtained therefrom. The interlocking feature extraction step S323 includes a driver pupil interlocking feature extraction step S3231 and a driver zone interlocking feature extraction step S3233, as shown in FIG. 16, and the control unit in the driver pupil interlocking feature extraction step S3231. 20 extracts the pupil feature of the driver, that is, the pupil spacing, and the vertical distance between the horizontal distance Ho formed by the pupil gap and the line segment formed by the left and right pupils and the lip formed by the left and right pupils in step S3233. Extracts information about the area area formed by Vo.

Thereafter, the control unit 30 compares these driver linkage features with the driver features pre-stored in the storage unit 30 and performs the linkage feature comparison step S325 for comparing whether the corresponding driver is a suitable driver for the storage unit 30. Run The interlocking feature comparison step (S325) includes a horizontal size checking step (S3251), a horizontal size ratio calculating step (S3252), a horizontal size error determining step (S3253), a vertical size checking step (S3254), and a vertical size ratio calculation. A step S3255 and a vertical size error determination step S3256 are included. In the horizontal size checking step S3251, the control unit 20 checks the horizontal size of the driver zone interlocking feature, and compares the pupil gap of the driver's pupil feature among the interlocking features by comparing the pupil gap Ho of the certified driver's face. The ratio is calculated (S3252). The storage unit 30 includes horizontal size ratio preset data with respect to the horizontal size ratio. If the ratio calculated in step S3252 has a value outside the range of the horizontal size ratio preset data, an error in image acquisition or processing Is determined, and the control flow is switched to step S3257 to perform the warning function through the warning unit 51.

Further, in step S3254 and S3255, the vertical distance between the line segment formed by the pupil and the line segment formed by the lips obtained from the image according to the forward and backward movement of the certified driver, and the driver area characteristic among the driver features stored in the storage unit for the authenticated driver. Check the vertical size Vo and calculate it to calculate the vertical size ratio, and compare this with the vertical size ratio preset data for the vertical size ratio stored in the storage 30 to calculate a value outside this data range in step S3255. If the ratio is set, the control unit 20 determines that the error is detected in step S3256 and applies the warning control signal to the warning unit 51.

As described above, the angle of the side mirror corresponding to the interlocking feature identified in the interlocking feature comparison step S325 is calculated in the interlocking feature checking step S327. That is, the angle value of the appropriate side mirror at the position where the currently authenticated driver has moved the face forward and backward is calculated using the mirror interlocking angle state calculated in step S29. Thereafter, the controller 20 may apply the mirror motor control signal of the mirror motor unit corresponding to the angle value of the side mirror to the mirror motor unit to more stably maintain the driver's side and rear gaze.

The above embodiment is an example for describing the present invention, and the present invention may be modified in various ways using the driver's pupil feature, the driver's pupil color feature, and the driver's zone feature from the driver face image information. That is, the driver recognition apparatus and its control method of the present invention may be variously modified, such as being associated with a drowsiness alarm device.

10 ... Driver recognition device 20 ... Control unit
30 storage 40 operation
50 ... operating part

Claims (18)

A driver recognition device mounted on a vehicle,
A detector including an image sensor configured to detect an image of a driver's face;
An input unit operable by a passenger;
A control unit receiving a signal from the detection unit and the input unit;
A driver pupil feature indicating a driver pupil distance, a driver pupil color feature indicating a driver pupil color, and a driver zone feature indicating an area of an area formed by the driver pupil and the lips for determining recognition of a driver's face. A storage unit for storing driver characteristics;
And an operation unit connected to the control unit and configured to image the driver's face image according to an operation control signal of the control unit.
The controller recognizes the driver's suitability by comparing the driver's face image and the driver's features from the image sensor, and controls the operation of the operation unit provided in the vehicle. The method of claim 1,
Wherein,
A warning unit configured to execute a warning by applying a warning control signal from the control unit when the control unit is not authorized for the driver;
A vehicle starting unit which receives a vehicle starting control signal from the controller according to whether the controller verifies driver suitability;
A mirror motor unit configured to receive a vehicle side mirror control signal from the control unit according to whether the control unit verifies driver suitability;
And a seat motor unit configured to receive a vehicle seat control signal from the control unit according to whether the controller conforms to the driver suitability certification. The method of claim 1,
Wherein the input unit comprises:
A mirror switch for adjusting the direction of the side mirror of the vehicle,
A seat switch for adjusting the seat position of the vehicle,
A mirror interlock switch configured to execute a mirror interlocking mode for adjusting a direction of a side mirror of a vehicle according to a movement of a driver's face position using preset mirror data in the storage unit;
And a memory switch configured to store the position of the vehicle side mirror and the driver's seat adjusted by the mirror switch and the seat switch input by the driver to the storage unit. A detection unit mounted on the vehicle and including an image sensor detecting an image of a driver's face; An input unit operable by a passenger; A control unit receiving a signal from the detection unit and the input unit; A driver pupil feature indicating a driver pupil distance, a driver pupil color feature indicating a driver pupil color, and a driver zone feature indicating an area of an area formed by the driver pupil and the lips for determining recognition of a driver's face. A storage unit for storing driver characteristics; And a computing unit connected to the controller and configured to process the driver's face image according to the operation control signal of the controller, wherein the controller compares the driver's face image from the image sensor and the driver's feature to determine the suitability of the driver. Providing a driver recognition device for authenticating and controlling whether an operation unit provided in the vehicle is operated;
The wireless communication unit of the vehicle receives the key authentication information from the smart key possessed by the driver and transmits the key authentication information to the control unit, and the control unit executes the driver key authentication by comparing with the preset key information stored in the storage unit. Key authentication step,
A setting step of storing a driver feature for face recognition of the driver using the image information detected by the detection unit;
And a mode execution step of comparing the image information sensed by the detection unit with the driver feature stored in the storage unit and executing an operation mode according to a comparison result. 5. The method of claim 4,
The driver recognizing apparatus further includes a warning unit configured to execute a warning by applying a warning control signal from the control unit when the controller is not certified as a driver.
The key authentication step is:
A wireless key information communication step of receiving, by the wireless communication unit, key authentication information transmitted from the smart key;
A key information checking step of the control unit comparing the received key authentication information with the preset key information by applying a key information control signal to the operation unit;
And a key authentication warning step of executing a warning by applying a warning control signal to the warning part by the controller when the key authentication information and the preset key information do not coincide in the key information checking step. Recognition device control method. 5. The method of claim 4,
The setting step is:
A selection mode that is input between a setting authentication mode for inputting a driver characteristic through the input unit (memory switch) and a setting interlocking mode for setting an interlocking range of the vehicle side mirror by setting a forward and backward movement range of a position at which the driver moves on the vehicle seat. A setting mode input step of transmitting a signal to the controller;
A position adjustment input step of receiving an adjusted position signal of the vehicle seat through a seat switch of the input unit;
And a setting mode determination step of determining a setting mode input in the setting mode input step. The method according to claim 6,
If the setting mode determined in the setting mode determination step is a setting authentication mode, the step of performing a setting authentication mode for confirming and storing the driver characteristics from the face image information of the driver, characterized in that the driver recognition device control method. 8. The method of claim 7,
The setting authentication mode execution step is:
A setting authentication detection step of the control unit obtaining a face image of the driver by applying a detection control signal to the detection unit;
A setting image processing step of image-processing the driver's face image detected by the calculating unit in the setting authentication detecting step;
A setting authentication feature extraction step of extracting a driver region feature representing a driver pupil feature representing a driver's pupil gap, a driver's pupil color, an area of an area formed by the pupil and the lips of the driver, from the driver's face image processed by the calculator Wow,
And a setting authentication information storing step of storing the driver feature extracted in the setting authentication feature extracting step into the storage unit. The method according to claim 6,
If the setting mode determined in the setting mode determination step is a setting interlocking mode, the setting interlocking mode execution step of confirming and setting the interlocking range of the side mirror of the vehicle from the face image information of the driver in the front and rear movement area of the driver further comprises: Driver recognition device control method characterized in that. The method of claim 9,
The setting linked mode execution step is:
A forward movement checking step of acquiring a face movement image in the front movement region of the driver's face by the detection unit to check the front movement region;
A rearward movement checking step of acquiring a face movement image in the rearward movement region of the driver's face by the detection unit to identify the rearward movement region;
A face movement image processing step of performing image processing on the face movement image acquired in the forward movement confirming step and the backward movement confirming step;
A mirror angle calculating step of calculating mirror angles corresponding to the front and rear moving regions of the driver by using mirror data preset in the storage unit;
And a mirror interlocking angle state storing step of storing mirror interlocking angle states according to the front and rear movement regions of the driver calculated in the mirror angle calculating step in the storage unit. The method of claim 10,
The forward movement checking step is:
A forward movement instruction step of the control unit requesting a forward movement of the driver's face to the driver through the warning unit;
And a forward movement image acquisition step of acquiring a forward movement image of the driver by sensing the image of the driver's face when the driver moves forward,
The backward movement checking step is:
The rear movement instruction step of the control unit requesting a rear movement of the driver's face to the driver through the warning unit;
And a rearward moving image acquisition step of acquiring a rearward moving image of the driver by sensing the image of the driver's face when the driver moves backward. 5. The method of claim 4,
The mode execution step is:
An authentication mode in which the control unit compares the image information detected by the sensing unit with the driver features stored in the storage unit and authenticates the driver's face, and detects the authenticated driver's face image to position the vehicle side mirror according to the driver's movement. And an operation mode determination step of determining which of the operation modes to be executed among the operation modes including the mirror interlocking mode for adjusting the driver. 13. The method of claim 12,
If it is determined that the operation mode to be executed in the operation mode determination step, the authentication mode,
The authentication mode is:
An authentication detection step of acquiring a driver's face image;
An authentication image processing step of performing an image processing of the face image of the driver by the operation unit;
The operation unit may include a driver pupil authentication feature representing a driver's pupil gap from an image of a driver's face image processed, a driver pupil color authentication feature representing a driver pupil color, and a driver zone authentication feature representing an area of an area formed by the driver pupil and the lips An authentication feature extraction step of extracting a driver authentication feature;
A driver authentication feature comparison step of comparing the driver authentication feature with a driver feature stored in the storage unit;
An authentication feature checking step of confirming, by the controller, a feature compared in the driver authentication feature comparison step;
And an authentication data execution step of applying, by the control unit, an operation control signal to the operation unit, in accordance with a result of the confirmation in the authentication feature confirmation step. The method of claim 13,
The authentication feature extraction step is:
A driver pupil authentication feature extraction step of extracting a driver's pupil feature from the image of the driver's face processed by the image;
A driver pupil color authentication feature extraction step of extracting a driver's pupil color feature from the image of the driver's face processed by the image;
And a driver zone feature extraction step of extracting a driver zone authentication feature using the driver pupil authentication feature and the driver lips. The method of claim 13,
The step of comparing the authentication feature is:
An authentication pupil interval comparison step of comparing a driver pupil interval by comparing the driver pupil authentication feature and the driver pupil feature stored in the storage unit;
In the authentication pupil interval comparison step, the identification pupil interval equal determination step of checking whether the comparison result is the same;
A verification pupil color comparison step of comparing the driver pupil color characteristic by comparing the driver pupil color authentication characteristic and the driver pupil color feature stored in the storage unit;
In the authentication pupil color comparison step, the verification pupil color identical determination step of checking whether the comparison result is the same;
An authentication zone area comparison step of comparing the driver zone authentication feature with the driver zone feature stored in the storage unit to compare a driver zone area;
And determining the authentication zone area equality step of checking whether the comparison result is the same in the authentication zone area comparison step. 13. The method of claim 12,
If it is determined that the operation mode to be executed in the operation mode determination step is the interlocking mode,
The interlocking mode is:
An interlock detection step of acquiring a driver's face image;
A linked image processing step of performing image processing of the driver's face image by the operation unit;
Extracting a driver interlocking feature including a driver interlocking feature indicating an area of an area formed by the driver's pupils and lips and a driver's pupil linking feature indicating an area of a driver's pupil from the image of the driver's face processed by the operation unit; Wow,
A driver linking feature comparing step of comparing the driver linking feature and a driver feature stored in the storage unit;
An interlocking feature checking step of confirming, by the controller, a feature compared in the driver linking feature comparing step;
And a linkage data execution step of applying, by the control unit, an operation control signal to the operation unit, in accordance with the result of the check in the linkage feature identification step. 17. The method of claim 16,
The interworking feature extraction step is:
A driver pupil authentication feature extraction step of extracting a driver's pupil feature from the image of the driver's face processed by the image;
And a driver zone feature extracting step of extracting a driver zone linkage feature using the driver pupil linkage feature and the driver lips. 18. The method of claim 17,
The interworking feature comparison step is:
A horizontal size checking step of checking a horizontal size of the driver zone interlocking feature;
A horizontal size ratio calculating step of calculating a ratio of the horizontal size of the driver zone feature and the horizontal size of the driver zone interlocking feature among the authenticated driver features;
A horizontal size error determining step of comparing a horizontal size ratio error range with the horizontal size ratio among preset data preset and stored in the storage;
A vertical size checking step of checking a vertical size of the driver zone interlocking feature when it is determined that the horizontal size error is normal;
A vertical size ratio calculating step of calculating a ratio between the vertical size of the driver zone feature and the vertical size of the driver zone interlocking feature among the certified driver features;
And a vertical size error determination step of comparing a vertical size ratio error range and the vertical size ratio among preset data preset and stored in the storage unit.

Images