WO2002035452A1 - Eye image obtaining method, iris recognizing method, and system using the same - Google Patents

Eye image obtaining method, iris recognizing method, and system using the same Download PDF

Info

Publication number
WO2002035452A1
WO2002035452A1 PCT/KR2001/001797 KR0101797W WO0235452A1 WO 2002035452 A1 WO2002035452 A1 WO 2002035452A1 KR 0101797 W KR0101797 W KR 0101797W WO 0235452 A1 WO0235452 A1 WO 0235452A1
Authority
WO
WIPO (PCT)
Prior art keywords
iris
image
pupil
camera
user
Prior art date
Application number
PCT/KR2001/001797
Other languages
French (fr)
Inventor
Deunggu Lee
Original Assignee
Alpha Engineering Co., Ltd.
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Priority claimed from KR1020000062539A external-priority patent/KR100362484B1/en
Priority claimed from KR1020000069072A external-priority patent/KR100361141B1/en
Priority claimed from KR1020000070962A external-priority patent/KR20010016241A/en
Priority claimed from KR10-2000-0070964A external-priority patent/KR100376415B1/en
Application filed by Alpha Engineering Co., Ltd. filed Critical Alpha Engineering Co., Ltd.
Priority to AU2002211048A priority Critical patent/AU2002211048A1/en
Publication of WO2002035452A1 publication Critical patent/WO2002035452A1/en

Links

Classifications

    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
    • G06V40/00—Recognition of biometric, human-related or animal-related patterns in image or video data
    • G06V40/10—Human or animal bodies, e.g. vehicle occupants or pedestrians; Body parts, e.g. hands
    • G06V40/18—Eye characteristics, e.g. of the iris

Definitions

  • the present invention relates to an eye image obtaining method, an iris recognizing method and systems using the same.
  • a pupil boundary i.e., a boundary between an iris and a pupil
  • an iris boundary i.e., a boundary between an iris and the whites of one's eye
  • the human iris is surrounded by the pupil boundary and the iris boundary and has a donut shape.
  • the human iris includes tangled muscle tissues which control a size of the pupil. These muscle tissues form an inherent pattern of an individual and also show a genetic, constitutional feature of an individual.
  • a technology using the human iris is classified into an iridology and an iris recognition technology.
  • the iridology is usually used in medical field such as the oriental medicine.
  • a constitution or a health state of a person is diagnosed by detecting a shape and a color of the iris and a shape of the pupil.
  • the iris recognition technology is based on the fact that the iris patterns of individuals differ.
  • Such an iris recognition technology identifies a person using an iris recognition, and is employed in various industrial fields such as a security system or an identification system.
  • Korean Patent Application no. 98-57393 discloses an iris recognition system.
  • the iris recognition system of Korean Patent Application no. 98-57393 includes a CCD module for converting an image input through a lens into a digital signal, and zooming and focusing an image desired to be recognized according to a control signal; a first signal processor for processing the digital signal applied from the CCD module; a second signal processor for calculating a size of the image desired to be recognized according to the signal processed by the first signal processor; and a controller for calculating a direction and a distance of the image desired to be recognized according to the signal calculated by the second signal processor to control a zoom operation and a focusing operation of the CCD module.
  • the iris recognition system obtains an iris image as follows: first, a size of an iris obtained by the CCD module is calculated. A distance and a direction of from the CCD module to a person standing in front of the iris recognition system are calculated. According to the calculation results, a camera of the CCD module is zoomed and focused. Thereafter, the iris of the person is photographed by the CCD module, whereby the iris image is obtained. Consequently, there is an inconvenience in that a user has to wait for a long time. Also, when the iris recognition system cannot recognize the obtained iris image, a user has to wait for a long time until the iris image is obtained again, thereby increasing an inconvenience. In addition, since the iris of the user is not photographed at a fixed location, the iris recognition system recognizes the iris image obtained by photographing the iris of the user moving, thereby lowering a recognition rate of the iris recognition system.
  • Korean Patent Application no. 1999-418 discloses an iris recognition system.
  • the iris recognition system of Korean Patent Application no. 1999-418 includes an inside case in which a camera for photographing an iris of a user, a lighting apparatus for providing light for the camera to photograph the iris image regardless of ambient brightness, and a distance for measuring sensor for measuring a distance from the iris are supported.
  • the iris recognition system obtains an iris image such that a distance from the iris is measured by the distance measuring sensor when an optical axis of the iris coincides with that of the camera by rotating the inside case, and the camera is zoomed and focused by the measured distance. Therefore, there are disadvantages that a time to obtain the iris image is lengthy, and a high cost automatic focusing camera is used.
  • preferred embodiments of the present invention provide an iris image obtaining method which is convenient.
  • the preferred embodiments of the present invention provide a method of obtaining an eye image, comprising: a) initializing a photographing unit; b) setting a reference image having a predetermined size in a central portion of an image input through a lens of the photographing unit; c) setting an allowable range of the reference image; d) matching an eye of a user with a focus axis of the photographing unit when the image input through the photographing unit deviates from the allowable range of the reference image; e) photographing the image input through the photographing unit; and f) detecting the eye image among the photographed images; and g) determining whether the eye image is a focused image.
  • the step (b) includes calculating a gray level of each of pixels forming the reference image and a dispersion value of the reference by using a histogram dispersion.
  • the allowable range of the reference image is set by a maximum value and a minimum value of the dispersion value.
  • the step (g) is performed by using a definition of a glint reflected from the eye image and an arrangement of an image of a lighting unit in a glint.
  • the step (f) is performed by a using matching degree between a pupil of the eye image and a circular template.
  • the present invention further provides a method of recognizing an iris, comprising: a) detecting a substantial pupil boundary by analyzing a histogram dispersion of an obtained iris image; b) forming a pupil replica boundary; c) calculating a distortion of a pupil by a difference between the substantial pupil boundary and the replica boundary; d) detecting an iris boundary; e) detecting an iris pattern comprised of the pupil boundary and the iris boundary; and f) determining whether the obtained iris image is a left iris or a right iris.
  • the step (a) is performed by a method of binarizing the iris image.
  • the step (a) includes detecting a first maximum point and a first minimum point when the histogram dispersion is observed from a dark region to a bright region; and separating a pupil region from an iris region by setting a brightness value at the first minimum point as a critical value.
  • the replica boundary is calculated by calculating a central point and a radius of the pupil replica.
  • the step (f) is performed by analyzing a location of the central point of the pupil with respect to a location of the central point of the iris.
  • the method further comprises, before the step (a), determining whether a user wears glasses, wherein the histogram dispersion becomes narrower when the user wears glasses.
  • the present invention further provides an eye image obtaining system, comprising: a translucent mirror arranged on a front surface of the eye image obtaining system and including an iris matching point; a camera arranged at a location corresponding to the iris matching point to photograph an iris image of a user; a lighting unit for throwing a light on a focal portion of the camera; and an indication lamp indicating that the camera is focused on the iris image.
  • the lighting unit includes a plurality of infrared light-emitting diodes arranged at a regular interval, and the lighting unit is tilted by a predetermined degree to emit an infrared ray toward the iris of the user.
  • a clear iris image is obtained by analyzing a definition of a glint reflected from the iris and an arranged shape of the light-emitting diodes.
  • the camera is a fixed focus camera, and the camera is focused on the iris image such that the user looks at the iris mathching point and moves forward.
  • the present invention further provides an entrance door system, comprising: a sensor portion detecting an approach of a person to generate a detecting signal; an iris image obtaining portion for obtaining an iris image; a memory portion storing registered iris images; a comparator comparing the obtained iris image with the registered iris image to a comparing signal; a controller driving the entrance door system in response to the detecting signal of the sensor portion and outputting a control signal in response to the comparing signal of the comparator; and an entrance door actuator opening or closing an entrance door in response to the control signal of the controller.
  • the entrance door system further comprises a digital processing portion converting the iris image obtained by the iris image obtaining portion into a digital signal; a key pad portion including key buttons to input a digit or a letter; a voice/text processing portion outputting a corresponding voice message or a con-esponding text message in response to the control signal output from the controller; and a call portion including a call button to a call a person inside the entrance door.
  • An iris recognition of the entrance door system is performed by calculating a gray level of the obtained iris image to detect a location of a glint reflected from the iris image, determining whether a dark region exists in the glint region or not, and detecting an image of the lighting means formed on the dark region of the iris image to determine whether the light-emitting diodes of the image are arranged at a regular interval or not.
  • the iris eye image obtaining portion includes: a translucent mirror arranged on a front surface of the iris image obtaining portion and including an iris matching point; a camera arranged at a location corresponding to the iris matching point to photograph an iris image of a user; a lighting unit for throwing a light on a focal portion of the camera; and an indication lamp indicating that the camera is focused on the iris image.
  • the lighting unit includes a plurality of infrared light-emitting diodes arranged at a regular interval, and the lighting unit is tilted by a predetermined degree to emit an infrared ray toward the iris of the user.
  • a clear iris image is obtained by analyzing a definition of a glint reflected from the iris and an arranged shape of the light-emitting diodes.
  • the camera is a fixed focus camera, and the camera is focused on the iris image such that the user looks at the iris matching point and moves forward.
  • the obtaining process is simplified, whereby increasing a convenience of a user. Also, using the iris image obtaining method according to the present invention, a clear iris image can be obtained. Further, using the iris recognition method according to the present invention, a high quality iris recognition can be achieved. Furthermore, when the iris recognition system using the iris image obtaining method and the iris recognition method is employed in various industrial filed, for example, a security system and a medical field, a high security system and an accurate diagnosis can be achieved.
  • FIG. 1 is a schematic block diagram illustrating an eye image obtaining system according to the present invention
  • FIG. 2 is a photograph illustrating an initial image after initializing a photographing unit according to the present invention
  • FIG. 3 is a photograph illustrating a reference image range according to the present invention
  • FIG. 4 shows a method of determining a definition of the eye image
  • FIG. 5 is a photograph illustrating a clear eye image obtained according to the present invention
  • FIG. 6 is a flow chart illustrating a method of obtaining an eye image according to the present invention
  • FIG. 7 is a photograph illustrating an iris image photographed under an infrared ray environment according to the present invention.
  • FIG. 8A shows a histogram dispersion of the iris image
  • FIG. 8B shows a pupil region separated from an iris region according to the present invention
  • FIG. 9A shows the iris image when a user wears glasses
  • FIG. 9B shows the histogram dispersion when a user wears glasses
  • FIG. 9C shows the pupil region separated from the iris region when the user wears glasses
  • FIG. 10A shows a pupil replica to obtain a central point of a pupil according to the present invention
  • FIG. 10B shows a difference between a pupil replica boundary and a substantial pupil boundary
  • FIG. 11A shows a gaze direction of human eyes
  • FIG. 11B shows a location of a central point of the pupil with respect to a location of a central point of the iris;
  • FIG. 12 is a flow chart illustrating an iris recognition method according to the present invention.
  • FIG. 13 shows an iris image obtaining system according to the present invention.
  • FIG. 14 shows a light means according to the present invention
  • FIG. 15 is a photograph illustrating the iris image obtained by the iris image obtaining system of FIG. 13;
  • FIG. 16 is a flow chart illustrating operation of the iris image obtaining system of FIG. 13
  • FIG. 17 is a block diagram illustrating an entrance door system according to the present invention.
  • FIG. 18 shows an iris image obtaining portion of the entrance door system of FIG. 17
  • FIG. 19 shows an appearance of the entrance door system of FIG. 17
  • FIG. 20 is a flow chart illustrating operation of the entrance door system of FIG. 17.
  • FIGs. 1 to 5 show a method of obtaining an eye image according to the present invention.
  • FIG. 1 is a schematic block diagram illustrating an eye image obtaining system according to the present invention.
  • the eye image obtaining system of FIG. 1 includes a photographing unit 1, a lighting unit 2 and a microcomputer 3.
  • the photographing unit 1 photographs an eye image in real-time.
  • a fixed focus camera, a digital camera, an image camera, or a video camera is used as the photographing unit 1.
  • the lighting unit 2 includes a plurality of light-emitting elements arranged at a regular interval, and throws a light on a focusing portion of the photographing unit 1 when an approach of a user is detected, so that an eye image can be photographed regardless of ambient brightness.
  • the microcomputer 3 includes a storing means (not shown) such as a buffer which stores eye images photographed by the photographing unit 1 and a reset button (not shown) which initializes the photographing unit 1.
  • a storing means such as a buffer which stores eye images photographed by the photographing unit 1
  • a reset button (not shown) which initializes the photographing unit 1.
  • FIG. 2 is a photograph illustrating an initial image after initializing a photographing unit according to the present invention.
  • the reference image stands for a minimum range which can detect a variation of an image photographed in real-time by a lens of the photographing unit 1.
  • FIG. 3 is a photograph illustrating the reference image range according to the present invention.
  • the microcomputer 3 regards an approach of the user as an intentional approach to output a message to match the eye of the user with a focus axis of the photographing unit 1.
  • the message includes a text message and a voice message.
  • the text message is output through a display means such as a liquid crystal display (LCD), and a voice message is output through a voice output means such as a speaker.
  • LCD liquid crystal display
  • a circular template is used in order to detect a pupil regardless of a kind of an image when the eye image exists. In other words, it is determined by a matching degree between the circular template and a pupil whether a pupil exists in the eye image.
  • Such a circular template is programmed in a storing means.
  • the microcomputer 3 determines whether the input eye image is a focused image or not by using a definition of a glint (see 15 in FIG. 7) reflected from the eye image and a definition of an image of the light emitting elements arranged at a regular interval in a glint.
  • a definition of a glint is classified from 0 to 255.
  • the present invention determines a definition of the eye image by the number of pixels in which a definition of a glint is more than 253.
  • FIG. 4 shows a method of determining a definition of the eye image. As a result, as shown in FIG. 5, the microcomputer 3 can obtain a clear eye image.
  • FIG. 5 is a photograph illustrating a clear eye image obtained according to the present invention.
  • FIG. 6 is a flow chart illustrating a method of obtaining an eye image according to the present invention.
  • the photographing unit 1 is initialized (step S102).
  • the microcomputer 3 sets a reference image having a predetermined size in a central portion of an image input through a lens of the photographing unit 1.
  • a dispersion value of each of several frames of set images is calculated by using a histogram dispersion.
  • An allowable range of the reference image is set by a maximum value and a minimum value of the dispersion value (step S 104).
  • the microcomputer 3 regards an approach of the user as an intentional approach (step S106). Hence, the microcomputer 3 outputs a message that asks to match an eye of the user with a focus axis of the photographing unit 1 (step S108).
  • the photographing unit 1 photographs an image in real-time, and the photographed images are stored in a storing means such as a buffer. Thereafter, it is determined whether an eye image exists in the photographed images (step SI 10). When an eye image does not exist in the photographed images stored in the storing means, the step S110 is repeatedly performed until an eye image is detected.
  • the microcomputer 3 determines whether the input eye image is a focused image or not by using a definition of a glint reflected from the eye image and a definition of an image of the light emitting elements arranged at a regular interval in a glint (step S 112).
  • FIGs. 7 to 11B show a method of recognizing an iris according to the present invention.
  • a method of binarizing a photographed iris image is used to divide only a pupil region from the photographed iris image. That is, a binarization method is used to detect a pupil boundary.
  • a method of detecting the pupil boundary using the binarization method is much faster in data processing speed than a conventional method using a black and white image. Therefore, it is possible to process the iris image in real-time. Also, since all pixels on the pupil boundary are detected, a substantial pupil boundary can be detected.
  • it is required to binarize the iris image by setting an appropriate critical value. Such a critical value depends on an obtained iris image, and thus a method of automatically setting a critical value with respect to an iris images is required.
  • FIG. 7 is a photograph illustrating the iris image photographed under an infrared ray environment according to the present invention.
  • FIG. 8A shows a histogram dispersion of the iris image.
  • the pupil region can be separated from the iris region by setting a brightness value at the minimum point 22 as a critical value.
  • the brightness values at the minimum points are connected to form a substantial pupil boundary. Therefore, the iris image can be binarized in such a way that, for example, a pupil region is set to "0" and an iris region is set to "1".
  • the pupil boundary can be detected rapidly.
  • An iris boundary can also be detected by the same method. Therefore, an iris pattern including the pupil boundary and the iris boundary can be detected rapidly.
  • the iris pattern can be used, for example, to identify a person.
  • FIG. 8B shows the pupil region separated from the iris region according to the present invention.
  • FIGs. 9A and 9B show the iris image and the histogram dispersion when a user wears glasses, respectively.
  • the histogram dispersion of FIG. 9B is narrower than that of FIG. 8B, and the first minimum point 22 of FIG. 9B is heightened. This is because a definition of the iris image is reduced due to a light reflection from a spectacle lens. Therefore, it can be discriminated by analyzing this histogram dispersion whether a user wears glasses or not. In order to obtain a clear iris image, a message that asks to take off the glasses can be output.
  • FIG. 9A and 9B show the iris image and the histogram dispersion when a user wears glasses, respectively.
  • the histogram dispersion of FIG. 9B is narrower than that of FIG. 8B, and the first minimum point 22 of FIG. 9B is heightened. This is because a definition of the iris image is reduced due to a light reflection from a spectacle lens. Therefore, it can be discriminated by
  • FIG. 10A shows a pupil replica to obtain a central point of the pupil.
  • the central point of the pupil is obtained as follows: an imaginary horizontal line 23 and an imaginary vertical line 24 perpendicular to the line 23 are drawn to pass through the pupil replica.
  • Boundary points 23a and 23b which are crossing points between a boundary of the pupil replica and the imaginary horizontal line 23 are obtained.
  • Boundary points 24a and 24b which are crossing points between a boundary of the pupil replica and the imaginary vertical line 24 are obtained.
  • a middle point 25 of the boundary points 23a and 23b and a middle point 26 of the boundary points 24a and 24b are obtained. Obtained is a crossing point that a line vertically extending from the middle point 25 crosses a line horizontally extending from the middle point 26. In the same way, an imaginary horizontal line and an imaginary vertical line are drawn, and several crossing points are obtained.
  • the central point of the pupil is an average value of the several crossing points.
  • FIG. 10B shows a difference between a pupil replica boundary and a substantial pupil boundary.
  • a reference numeral 37 denotes the substantial pupil boundary
  • a reference numeral 38 denotes the pupil replica boundary.
  • a reference numeral 39 denotes a distortion between the substantial pupil boundary 37 and the replica boundary 38.
  • a distortion of the pupil can be analyzed by a difference between the substantial pupil boundary 37 and the replica boundary 38.
  • a distortion of the pupil can be used to diagnose, for example, a health state of a person.
  • FIG. UA shows a gaze direction of human eyes.
  • arrows denote a gaze direction of human eyes.
  • a gaze direction of human eyes is turned on the middle.
  • the gaze direction affects a location of the central point of the pupil. Therefore, as shown in FIG. 11B, the central point of the left pupil is located on the left side of the central point of the left iris, and the central point of the right pupil is located on the right side of the central point of the right iris.
  • 11B shows a location of a central point of the pupil with respect to a location of a central point of the iris. Therefore, when the central point 27 of the pupil and the central point 28 of the iris are accurately detected, it is possible to discriminate whether the obtained iris image is the left iris or the right iris.
  • FIG. 12 is a flow chart illustrating an iris recognition method according to the present invention.
  • the substantial pupil boundary is detected by analyzing a histogram dispersion of an obtained iris image (step S130). That is, a pupil region is separated from the iris region by setting a brightness value at the first minimum point of the histogram dispersion as a critical value. Thereafter, the replica boundary is calculated by calculating a central point and a radius of the pupil replica (step SI 32). A distortion of the pupil is calculated by a difference between the substantial pupil boundary and the replica boundary (step S134). Thereafter, an iris boundary is detected (step S136). Then, an iris pattern comprised of the pupil boundary 12 and the iris boundary 13 is detected (step S138). It is discriminated by analyzing a location of the central point of the pupil with respect to a location of the central point of the iris whether the obtained iris image is the left iris or the right iris (step S140).
  • step S130 when a histogram dispersion is relatively narrow, it is determined that a user wears glasses. Therefore, a step for asking to take off the glasses can be added after the step 130. Otherwise, an algorithm for a user wearing glasses can be added.
  • the eye image obtaining method and the iris recognition method can be employed in various industrial fields.
  • FIG. 13 shows an iris image obtaining system according to the present invention.
  • a translucent mirror 40 is arranged on a front surface of the iris image obtaining system 90.
  • the translucent mirror 40 includes an iris matching point 42.
  • a fixed focus camera 44 is arranged at a location corresponding to the iris matching point 42 to photograph an iris image of a user. Therefore, when a user looks at the iris matching point
  • the iris image obtaining system 90 further includes a lighting means 50 in order to obtain the iris image regardless of a brightness of ambient light.
  • FIG. 14 shows the light means 50 according to the present invention. As shown in FIG.
  • the lighting means 50 includes a plurality of infrared light-emitting diodes 52 arranged at a regular interval in a cross form.
  • the lighting means 50 is tilted by a predetermined degree to emit an infrared ray toward the iris of the user.
  • FIG. 15 is a photograph illustrating the iris image obtained by the iris image obtaining system of FIG. 13. As shown in FIG. 15, a definition of a glint reflected from the iris and an arranged shape of the light-emitting diodes are analyzed, whereupon a clear iris image can be obtained. Operation of the iris image obtaining system of FIG. 13 is described below with reference to FIG. 16. FIG. 16 is a flow chart illustrating operation of the iris image obtaining system of FIG. 13.
  • step 162 when a user looks at the iris matching point 48 of the translucent mirror 40, it is detected whether an iris image exists in a photographed image or not (step 162).
  • the iris images are photographed in real-time during a forward movement of the user to focus the camera 40 on the iris of the user (step S164).
  • a location of a glint reflected from the iris image is detected by calculating a gray level of the obtained iris image (step SI 66). Thereafter, it is determined whether a dark region exists in the glint region or not (step S168). An image of the lighting means formed on the dark region of the iris image is detected (step S170). And, it is discriminated whether the light-emitting diodes of the lighting mean image are arranged at a regular interval or not (step S172). When the light-emitting diodes of the lighting mean image are not arranged at a regular interval and are vague, the process comes back to the step S168. When the light-emitting diodes of the lighting mean image are arranged at a regular interval and are clear, the obtained iris image is discriminated as a clear iris image (step S174).
  • FIG. 17 is a block diagram illustrating an entrance door system according to the present invention.
  • the entrance door system of FIG. 17 employs the eye image obtaining method and the iris recognition method described above.
  • FIG. 18 shows an iris image obtaining portion of the entrance door system of FIG. 17.
  • FIG. 19 shows an appearance of the entrance door system of FIG. 17.
  • the entrance door system of FIG. 17 includes a sensor portion 54, a key pad portion 56, a call portion 58, a voice/text processing portion 60, a memory portion 67, a buffer portion 69, a digital processing portion 70, a comparator 72, a controller 74, a video camera 76, a video phone 78, an entrance door actuator 80, and an iris image obtaining portion 90.
  • the sensor portion 54 detects an approach of a user and outputs a detecting signal to the controller 74.
  • the key pad portion 56 includes key buttons and is used to input a digit or a letter. For example, a password or an identification (ID) number can be input through the key pad portion 56.
  • the call portion 58 includes a call button (not shown) and is used to a call a person (e.g., staff member of a company) inside an entrance door.
  • the voice/text processing portion 60 includes a voice processing portion 62, a voice output portion 64, a memory portion 65, and a display portion 66.
  • the voice/text processing portion 60 outputs a corresponding voice message or a corresponding text message in response to a control signal output from the controller 74 through the voice output portion 64 or the display portion 66.
  • the memory portion 65 stores various voice/text messages.
  • the memory portion 67 stores various data required for an iris recognition such as registered iris images, a user information, and a password.
  • the digital processing portion 70 converts the input iris image into a digital signal and stores it in the buffer portion 69 temporarily.
  • the comparator 72 compares the input iris image with the registered iris image stored in the memory portion 67 and outputs a comparing signal to the controller 74 to determine whether a user is a person allowed to enter the entrance door or not.
  • the controller 74 controls all components of the entrance door system of FIG. 13.
  • the controller 74 controls the iris image obtaining portion 90 to obtain a clear iris image.
  • the controller 74 calculates a gray level of the input iris image, and detects a location of a glint reflected from the iris image. Then, the controller 74 determines whether a dark region exists in the glint region or not, and detects an image of the lighting means formed on the dark region of the iris image to determine whether the light-emitting diodes of the image are arranged at a regular interval or not. When the light- emitting diodes of the image are arranged at a regular interval and are clear, the controller 74 regards the input iris image as a clear iris image.
  • the person inside the entrance door can monitor the user through the video camera 76 and the video phone 78.
  • the entrance door actuator 80 opens or closes the entrance door in response to a control signal output from the controller 74.
  • the iris image obtaining portion 90 of FIG. 18 has the same configuration as the eye image obtaining system of FIG. 13. Like reference numerals of FIGs. 13 and 18 denote like parts, and thus their description is omitted to avoid a redundancy. That is, the iris image obtaining portion 90 obtains an iris image by the method of FIG. 16.
  • FIG. 20 is a flow chart illustrating operation of the entrance door system of FIG. 17.
  • the sensor portion 54 detects an approach of the user and outputs a detecting signal to the controller 74 (step S182).
  • the controller 74 outputs a corresponding voice message or a conesponding text message stored in the memory portion 65 through the voice/text processing portion 60 in response to the detecting signal (step SI 84).
  • step SI86 it is determined whether the user pushes the call button of the call portion 58 in a predetermined time period or not.
  • the controller 74 outputs a call-related voice message or a call-related text message before the entrance door system is initialized (step S188).
  • the controller 74 When the user does not the call button of the call portion 58 in a predetermined time period, the controller 74 outputs a voice message or a text message that asks to input an iris of the user. Hence, the user looks at the iris matching point 42 of the translucent mirror 40 to match the iris with a focal axis of the fixed focus camera 44, and then moves forward to focus the camera 44 on the iris. During a forward movement to focus the camera 44 on the iris, the fixed focus camera 44 photographs the iris images in real-time (step S 190).
  • the photographed iris images are converted into the digital signals by the digital processing portion 70 and then are temporarily stored in the buffer portion 69.
  • the controller 74 calculates a gray level of the input iris image, and detects a location of a glint reflected from the iris image (step S192). Then, the controller 74 determines whether a dark region exists in the glint or not (step S194). When a dark region exists in the glint, the controller 74 also detects an image of the lighting means formed on the dark region of the iris image to determine whether the light-emitting diodes are arranged at a regular interval or not (step S196). When the light-emitting diodes are ananged at a regular interval and are clear, the controller 74 regards the input iris image as a clear iris image (step S 198).
  • the controller 74 When the clear iris image is obtained, the controller 74 outputs a voice message or a text message through the voice output portion 64 or the display portion 66 in order to notify the user that the input iris image is recognized (step S200).
  • the comparator 72 determines whether the registered iris image stored in the memory portion 67 is identical to the obtained iris image or not (step S202).
  • the controller 74 outputs a control signal to the entrance door actuator 80 to open the entrance door (step S204).
  • the controller 74 determines whether a predetermined time period is passed or not (step S206). After a predetermined time period is passed, the controller outputs a control signal to the entrance door actuator 80 to close the entrance door (step S208).
  • the obtaining process is simplified, whereby increasing a convenience of a user.
  • a clear iris image can be obtained.
  • a high quality iris recognition can be achieved.
  • the iris recognition system using the iris image obtaining method and the iris recognition method is employed in various industrial filed, for example, a security system and a medical field, a high security system an accurate diagnosis can be achieved.

Abstract

A method of obtaining an eye image includes a) initializing a photographing unit; b) setting a reference image having a predetermined size in a central portion of an image input through a lens of the photographing unit; c) setting an allowable range of the reference image; d) matching an eye of a user with a focus axis of the photographing unit when the image input through the photographing unit deviates from the allowable range of the reference image; e) photographing the image input through the photographing unit; and f) detecting the eye image among the photographed images; and g) determining whether the eye image is a focused image.

Description

EYE IMAGE OBTAINING METHOD, IRIS RECOGNIZING METHOD, AND
SYSTEM USING THE SAME
Technical Field The present invention relates to an eye image obtaining method, an iris recognizing method and systems using the same.
Background Art
In general, in order to use a human iris as information, a pupil boundary (i.e., a boundary between an iris and a pupil) and an iris boundary (i.e., a boundary between an iris and the whites of one's eye) have to be detected. The human iris is surrounded by the pupil boundary and the iris boundary and has a donut shape. The human iris includes tangled muscle tissues which control a size of the pupil. These muscle tissues form an inherent pattern of an individual and also show a genetic, constitutional feature of an individual.
A technology using the human iris is classified into an iridology and an iris recognition technology. The iridology is usually used in medical field such as the oriental medicine. In the iridology, a constitution or a health state of a person is diagnosed by detecting a shape and a color of the iris and a shape of the pupil. The iris recognition technology is based on the fact that the iris patterns of individuals differ. Such an iris recognition technology identifies a person using an iris recognition, and is employed in various industrial fields such as a security system or an identification system.
Korean Patent Application no. 98-57393 (Publication no. 2000-41509) discloses an iris recognition system. The iris recognition system of Korean Patent Application no. 98-57393 includes a CCD module for converting an image input through a lens into a digital signal, and zooming and focusing an image desired to be recognized according to a control signal; a first signal processor for processing the digital signal applied from the CCD module; a second signal processor for calculating a size of the image desired to be recognized according to the signal processed by the first signal processor; and a controller for calculating a direction and a distance of the image desired to be recognized according to the signal calculated by the second signal processor to control a zoom operation and a focusing operation of the CCD module.
The iris recognition system obtains an iris image as follows: first, a size of an iris obtained by the CCD module is calculated. A distance and a direction of from the CCD module to a person standing in front of the iris recognition system are calculated. According to the calculation results, a camera of the CCD module is zoomed and focused. Thereafter, the iris of the person is photographed by the CCD module, whereby the iris image is obtained. Consequently, there is an inconvenience in that a user has to wait for a long time. Also, when the iris recognition system cannot recognize the obtained iris image, a user has to wait for a long time until the iris image is obtained again, thereby increasing an inconvenience. In addition, since the iris of the user is not photographed at a fixed location, the iris recognition system recognizes the iris image obtained by photographing the iris of the user moving, thereby lowering a recognition rate of the iris recognition system.
Also, Korean Patent Application no. 1999-418 (Publication no. 2000-50494) discloses an iris recognition system. The iris recognition system of Korean Patent Application no. 1999-418 includes an inside case in which a camera for photographing an iris of a user, a lighting apparatus for providing light for the camera to photograph the iris image regardless of ambient brightness, and a distance for measuring sensor for measuring a distance from the iris are supported.
The iris recognition system obtains an iris image such that a distance from the iris is measured by the distance measuring sensor when an optical axis of the iris coincides with that of the camera by rotating the inside case, and the camera is zoomed and focused by the measured distance. Therefore, there are disadvantages that a time to obtain the iris image is lengthy, and a high cost automatic focusing camera is used.
Disclosure of Invention
To overcome the problems described above, preferred embodiments of the present invention provide an iris image obtaining method which is convenient.
It is another object of the present invention to provide an iris recognition method which can achieve a high quality iris recognition.
In order to achieve the above object, the preferred embodiments of the present invention provide a method of obtaining an eye image, comprising: a) initializing a photographing unit; b) setting a reference image having a predetermined size in a central portion of an image input through a lens of the photographing unit; c) setting an allowable range of the reference image; d) matching an eye of a user with a focus axis of the photographing unit when the image input through the photographing unit deviates from the allowable range of the reference image; e) photographing the image input through the photographing unit; and f) detecting the eye image among the photographed images; and g) determining whether the eye image is a focused image.
The step (b) includes calculating a gray level of each of pixels forming the reference image and a dispersion value of the reference by using a histogram dispersion. The allowable range of the reference image is set by a maximum value and a minimum value of the dispersion value. The step (g) is performed by using a definition of a glint reflected from the eye image and an arrangement of an image of a lighting unit in a glint. The step (f) is performed by a using matching degree between a pupil of the eye image and a circular template. The present invention further provides a method of recognizing an iris, comprising: a) detecting a substantial pupil boundary by analyzing a histogram dispersion of an obtained iris image; b) forming a pupil replica boundary; c) calculating a distortion of a pupil by a difference between the substantial pupil boundary and the replica boundary; d) detecting an iris boundary; e) detecting an iris pattern comprised of the pupil boundary and the iris boundary; and f) determining whether the obtained iris image is a left iris or a right iris.
The step (a) is performed by a method of binarizing the iris image. The step (a) includes detecting a first maximum point and a first minimum point when the histogram dispersion is observed from a dark region to a bright region; and separating a pupil region from an iris region by setting a brightness value at the first minimum point as a critical value. The replica boundary is calculated by calculating a central point and a radius of the pupil replica. The step (f) is performed by analyzing a location of the central point of the pupil with respect to a location of the central point of the iris. The method further comprises, before the step (a), determining whether a user wears glasses, wherein the histogram dispersion becomes narrower when the user wears glasses.
The present invention further provides an eye image obtaining system, comprising: a translucent mirror arranged on a front surface of the eye image obtaining system and including an iris matching point; a camera arranged at a location corresponding to the iris matching point to photograph an iris image of a user; a lighting unit for throwing a light on a focal portion of the camera; and an indication lamp indicating that the camera is focused on the iris image.
The lighting unit includes a plurality of infrared light-emitting diodes arranged at a regular interval, and the lighting unit is tilted by a predetermined degree to emit an infrared ray toward the iris of the user. A clear iris image is obtained by analyzing a definition of a glint reflected from the iris and an arranged shape of the light-emitting diodes. The camera is a fixed focus camera, and the camera is focused on the iris image such that the user looks at the iris mathching point and moves forward.
The present invention further provides an entrance door system, comprising: a sensor portion detecting an approach of a person to generate a detecting signal; an iris image obtaining portion for obtaining an iris image; a memory portion storing registered iris images; a comparator comparing the obtained iris image with the registered iris image to a comparing signal; a controller driving the entrance door system in response to the detecting signal of the sensor portion and outputting a control signal in response to the comparing signal of the comparator; and an entrance door actuator opening or closing an entrance door in response to the control signal of the controller.
The entrance door system further comprises a digital processing portion converting the iris image obtained by the iris image obtaining portion into a digital signal; a key pad portion including key buttons to input a digit or a letter; a voice/text processing portion outputting a corresponding voice message or a con-esponding text message in response to the control signal output from the controller; and a call portion including a call button to a call a person inside the entrance door.
An iris recognition of the entrance door system is performed by calculating a gray level of the obtained iris image to detect a location of a glint reflected from the iris image, determining whether a dark region exists in the glint region or not, and detecting an image of the lighting means formed on the dark region of the iris image to determine whether the light-emitting diodes of the image are arranged at a regular interval or not.
The iris eye image obtaining portion includes: a translucent mirror arranged on a front surface of the iris image obtaining portion and including an iris matching point; a camera arranged at a location corresponding to the iris matching point to photograph an iris image of a user; a lighting unit for throwing a light on a focal portion of the camera; and an indication lamp indicating that the camera is focused on the iris image.
The lighting unit includes a plurality of infrared light-emitting diodes arranged at a regular interval, and the lighting unit is tilted by a predetermined degree to emit an infrared ray toward the iris of the user. A clear iris image is obtained by analyzing a definition of a glint reflected from the iris and an arranged shape of the light-emitting diodes. The camera is a fixed focus camera, and the camera is focused on the iris image such that the user looks at the iris matching point and moves forward.
Using the iris image obtaining method according to the present invention, the obtaining process is simplified, whereby increasing a convenience of a user. Also, using the iris image obtaining method according to the present invention, a clear iris image can be obtained. Further, using the iris recognition method according to the present invention, a high quality iris recognition can be achieved. Furthermore, when the iris recognition system using the iris image obtaining method and the iris recognition method is employed in various industrial filed, for example, a security system and a medical field, a high security system and an accurate diagnosis can be achieved.
Brief Description of Drawings
For a more complete understanding of the present invention and the advantages thereof, reference is now made to the following descriptions taken in conjunction with the accompanying drawings, in which like reference numerals denote like parts, and in which: FIG. 1 is a schematic block diagram illustrating an eye image obtaining system according to the present invention;
FIG. 2 is a photograph illustrating an initial image after initializing a photographing unit according to the present invention; FIG. 3 is a photograph illustrating a reference image range according to the present invention;
FIG. 4 shows a method of determining a definition of the eye image; FIG. 5 is a photograph illustrating a clear eye image obtained according to the present invention; FIG. 6 is a flow chart illustrating a method of obtaining an eye image according to the present invention;
FIG. 7 is a photograph illustrating an iris image photographed under an infrared ray environment according to the present invention;
FIG. 8A shows a histogram dispersion of the iris image; FIG. 8B shows a pupil region separated from an iris region according to the present invention;
FIG. 9A shows the iris image when a user wears glasses; FIG. 9B shows the histogram dispersion when a user wears glasses; FIG. 9C shows the pupil region separated from the iris region when the user wears glasses;
FIG. 10A shows a pupil replica to obtain a central point of a pupil according to the present invention;
FIG. 10B shows a difference between a pupil replica boundary and a substantial pupil boundary; FIG. 11A shows a gaze direction of human eyes; FIG. 11B shows a location of a central point of the pupil with respect to a location of a central point of the iris;
FIG. 12 is a flow chart illustrating an iris recognition method according to the present invention.; FIG. 13 shows an iris image obtaining system according to the present invention.;
FIG. 14 shows a light means according to the present invention;
FIG. 15 is a photograph illustrating the iris image obtained by the iris image obtaining system of FIG. 13;
FIG. 16 is a flow chart illustrating operation of the iris image obtaining system of FIG. 13
FIG. 17 is a block diagram illustrating an entrance door system according to the present invention;
FIG. 18 shows an iris image obtaining portion of the entrance door system of FIG. 17; FIG. 19 shows an appearance of the entrance door system of FIG. 17; and
FIG. 20 is a flow chart illustrating operation of the entrance door system of FIG. 17.
Best Mode for Carrying Out the Invention Reference will now be made in detail to preferred embodiments of the present invention, example of which is illustrated in the accompanying drawings.
FIGs. 1 to 5 show a method of obtaining an eye image according to the present invention.
FIG. 1 is a schematic block diagram illustrating an eye image obtaining system according to the present invention. The eye image obtaining system of FIG. 1 includes a photographing unit 1, a lighting unit 2 and a microcomputer 3.
The photographing unit 1 photographs an eye image in real-time. A fixed focus camera, a digital camera, an image camera, or a video camera is used as the photographing unit 1. The lighting unit 2 includes a plurality of light-emitting elements arranged at a regular interval, and throws a light on a focusing portion of the photographing unit 1 when an approach of a user is detected, so that an eye image can be photographed regardless of ambient brightness.
The microcomputer 3 includes a storing means (not shown) such as a buffer which stores eye images photographed by the photographing unit 1 and a reset button (not shown) which initializes the photographing unit 1.
In a process of initializing the photographing unit 1, a reference image having a predetermined size is set in a central portion of an image which is input through a lens of the photographing unit 1 in a state that a user does not exist. A dispersion value is calculated using a histogram dispersion with respect to a gray level of the set reference image. An allowable range of the reference image used to detect an approach of the user is determined by a maximum value and a minimum value of the dispersion value. FIG. 2 is a photograph illustrating an initial image after initializing a photographing unit according to the present invention. The reference image stands for a minimum range which can detect a variation of an image photographed in real-time by a lens of the photographing unit 1. The present invention determines whether a user approaches or not by setting a pixel of 25x25 as a reference image. FIG. 3 is a photograph illustrating the reference image range according to the present invention. When a dispersion value of the reference image deviates from the allowable range during a predetermined time period due to an approach of a user, the microcomputer 3 regards an approach of the user as an intentional approach to output a message to match the eye of the user with a focus axis of the photographing unit 1. The message includes a text message and a voice message. The text message is output through a display means such as a liquid crystal display (LCD), and a voice message is output through a voice output means such as a speaker.
Thereafter, it is determined whether the eye image exists in an image input through the photographing unit 1 or not. In order to determine whether the eye image exists or not, it is detected whether a pupil exists in the eye image. A circular template is used in order to detect a pupil regardless of a kind of an image when the eye image exists. In other words, it is determined by a matching degree between the circular template and a pupil whether a pupil exists in the eye image. Such a circular template is programmed in a storing means.
Also, the microcomputer 3 determines whether the input eye image is a focused image or not by using a definition of a glint (see 15 in FIG. 7) reflected from the eye image and a definition of an image of the light emitting elements arranged at a regular interval in a glint.
When the eye image is vague, it is very difficult to find an image of the light emitting elements in the glint reflected from the eye image. Also, when the eye image is clear, it is very easy to find an image of the light emitting elements arranged at a regular interval in the glint reflected from the eye image
A definition of a glint is classified from 0 to 255. The present invention determines a definition of the eye image by the number of pixels in which a definition of a glint is more than 253. FIG. 4 shows a method of determining a definition of the eye image. As a result, as shown in FIG. 5, the microcomputer 3 can obtain a clear eye image.
FIG. 5 is a photograph illustrating a clear eye image obtained according to the present invention.
A method of obtaining an eye image according to the present invention is described below with reference to FIG. 6. FIG. 6 is a flow chart illustrating a method of obtaining an eye image according to the present invention.
First, the photographing unit 1 is initialized (step S102). The microcomputer 3 sets a reference image having a predetermined size in a central portion of an image input through a lens of the photographing unit 1. A dispersion value of each of several frames of set images is calculated by using a histogram dispersion. An allowable range of the reference image is set by a maximum value and a minimum value of the dispersion value (step S 104).
When an image input through the photographing unit 1 deviates from the allowable range of the reference image due to an approach of a user during a predetermined time period, the microcomputer 3 regards an approach of the user as an intentional approach (step S106). Hence, the microcomputer 3 outputs a message that asks to match an eye of the user with a focus axis of the photographing unit 1 (step S108). At the same time, the photographing unit 1 photographs an image in real-time, and the photographed images are stored in a storing means such as a buffer. Thereafter, it is determined whether an eye image exists in the photographed images (step SI 10). When an eye image does not exist in the photographed images stored in the storing means, the step S110 is repeatedly performed until an eye image is detected.
When an eye image is detected, the microcomputer 3 determines whether the input eye image is a focused image or not by using a definition of a glint reflected from the eye image and a definition of an image of the light emitting elements arranged at a regular interval in a glint (step S 112).
When the input eye image is a non-focused image, the microcomputer 3 outputs a message to notify the user that the input eye image is not recognized in order to make the user input an eye image again. When the input eye image is a focused image, the microcomputer 3 regards the input eye image as a clear eye image. Therefore, a clear eye image can be obtained (step S 114). FIGs. 7 to 11B show a method of recognizing an iris according to the present invention.
First, a method of binarizing a photographed iris image is used to divide only a pupil region from the photographed iris image. That is, a binarization method is used to detect a pupil boundary. A method of detecting the pupil boundary using the binarization method is much faster in data processing speed than a conventional method using a black and white image. Therefore, it is possible to process the iris image in real-time. Also, since all pixels on the pupil boundary are detected, a substantial pupil boundary can be detected. In order to divide the pupil from the iris, it is required to binarize the iris image by setting an appropriate critical value. Such a critical value depends on an obtained iris image, and thus a method of automatically setting a critical value with respect to an iris images is required. An infrared ray is used as a light environment to obtain the iris image. A division of the pupil from the iris is performed on the basis of a fact that the pupil 10 is lower in brightness than the iris 11, the whites 14, the eyelid 16, and the eyebrow 17. FIG. 7 is a photograph illustrating the iris image photographed under an infrared ray environment according to the present invention.
FIG. 8A shows a histogram dispersion of the iris image. As shown in FIG. 8A, when the histogram dispersion of FIG. 8 A is observed from a dark region to a bright region, a first maximum point 21 and a first minimum point 22 are found. The pupil region can be separated from the iris region by setting a brightness value at the minimum point 22 as a critical value. In more detail, the brightness values at the minimum points are connected to form a substantial pupil boundary. Therefore, the iris image can be binarized in such a way that, for example, a pupil region is set to "0" and an iris region is set to "1". The pupil boundary can be detected rapidly. An iris boundary can also be detected by the same method. Therefore, an iris pattern including the pupil boundary and the iris boundary can be detected rapidly. The iris pattern can be used, for example, to identify a person.
In order to obtain a wide pupil region, a value higher than a brightness value at the minimum point 22 can be set as a critical value. FIG. 8B shows the pupil region separated from the iris region according to the present invention.
Meanwhile, when a user wears glasses, the histogram dispersion becomes much narrower than when a user does not wear glasses. FIGs. 9A and 9B show the iris image and the histogram dispersion when a user wears glasses, respectively. The histogram dispersion of FIG. 9B is narrower than that of FIG. 8B, and the first minimum point 22 of FIG. 9B is heightened. This is because a definition of the iris image is reduced due to a light reflection from a spectacle lens. Therefore, it can be discriminated by analyzing this histogram dispersion whether a user wears glasses or not. In order to obtain a clear iris image, a message that asks to take off the glasses can be output. FIG. 9C shows the pupil region separated from the iris region when the user wears glasses. Meanwhile, a shape of the pupil depends on the pupil boundary, and the pupil boundary can be represented as a circular pupil replica having a central point and a radius. FIG. 10A shows a pupil replica to obtain a central point of the pupil. The central point of the pupil is obtained as follows: an imaginary horizontal line 23 and an imaginary vertical line 24 perpendicular to the line 23 are drawn to pass through the pupil replica. Boundary points 23a and 23b which are crossing points between a boundary of the pupil replica and the imaginary horizontal line 23 are obtained. Boundary points 24a and 24b which are crossing points between a boundary of the pupil replica and the imaginary vertical line 24 are obtained. A middle point 25 of the boundary points 23a and 23b and a middle point 26 of the boundary points 24a and 24b are obtained. Obtained is a crossing point that a line vertically extending from the middle point 25 crosses a line horizontally extending from the middle point 26. In the same way, an imaginary horizontal line and an imaginary vertical line are drawn, and several crossing points are obtained. The central point of the pupil is an average value of the several crossing points.
The radius is a length of a horizontal axis with respect to the central point. FIG. 10B shows a difference between a pupil replica boundary and a substantial pupil boundary. In FIG. 10B, a reference numeral 37 denotes the substantial pupil boundary, and a reference numeral 38 denotes the pupil replica boundary. Also, a reference numeral 39 denotes a distortion between the substantial pupil boundary 37 and the replica boundary 38. A distortion of the pupil can be analyzed by a difference between the substantial pupil boundary 37 and the replica boundary 38. A distortion of the pupil can be used to diagnose, for example, a health state of a person.
A discrimination between right and left iris images is performed by using a location of a central point of the pupil with respect to a location of a central point of the iris. FIG. UA shows a gaze direction of human eyes. In FIG. 11A, arrows denote a gaze direction of human eyes. As shown in FIG. 11 A, a gaze direction of human eyes is turned on the middle. The gaze direction affects a location of the central point of the pupil. Therefore, as shown in FIG. 11B, the central point of the left pupil is located on the left side of the central point of the left iris, and the central point of the right pupil is located on the right side of the central point of the right iris. FIG. 11B shows a location of a central point of the pupil with respect to a location of a central point of the iris. Therefore, when the central point 27 of the pupil and the central point 28 of the iris are accurately detected, it is possible to discriminate whether the obtained iris image is the left iris or the right iris.
A method of recognizing an iris according to the present invention is described below with reference to FIG. 12. FIG. 12 is a flow chart illustrating an iris recognition method according to the present invention.
First, the substantial pupil boundary is detected by analyzing a histogram dispersion of an obtained iris image (step S130). That is, a pupil region is separated from the iris region by setting a brightness value at the first minimum point of the histogram dispersion as a critical value. Thereafter, the replica boundary is calculated by calculating a central point and a radius of the pupil replica (step SI 32). A distortion of the pupil is calculated by a difference between the substantial pupil boundary and the replica boundary (step S134). Thereafter, an iris boundary is detected (step S136). Then, an iris pattern comprised of the pupil boundary 12 and the iris boundary 13 is detected (step S138). It is discriminated by analyzing a location of the central point of the pupil with respect to a location of the central point of the iris whether the obtained iris image is the left iris or the right iris (step S140).
In the step S130, when a histogram dispersion is relatively narrow, it is determined that a user wears glasses. Therefore, a step for asking to take off the glasses can be added after the step 130. Otherwise, an algorithm for a user wearing glasses can be added.
The eye image obtaining method and the iris recognition method can be employed in various industrial fields.
FIG. 13 shows an iris image obtaining system according to the present invention.
A translucent mirror 40 is arranged on a front surface of the iris image obtaining system 90. The translucent mirror 40 includes an iris matching point 42. A fixed focus camera 44 is arranged at a location corresponding to the iris matching point 42 to photograph an iris image of a user. Therefore, when a user looks at the iris matching point
42 of the translucent mirror 40, an iris of a user is matched with an optical axis of the camera 44 through the iris matching point 42. In order to focus the cameral 44 on the iris image, a user moves forward until an indication lamp 49 is seen through a slit 46 under the translucent mirror 40. In other words, when a user moves forward and sees the indication lamp 48 through the slit 46 after the iris is matched with an optical axis of the camera 44, the camera 44 is focused on the iris image. The iris image obtaining system 90 further includes a lighting means 50 in order to obtain the iris image regardless of a brightness of ambient light. FIG. 14 shows the light means 50 according to the present invention. As shown in FIG. 14, the lighting means 50 includes a plurality of infrared light-emitting diodes 52 arranged at a regular interval in a cross form. The lighting means 50 is tilted by a predetermined degree to emit an infrared ray toward the iris of the user.
FIG. 15 is a photograph illustrating the iris image obtained by the iris image obtaining system of FIG. 13. As shown in FIG. 15, a definition of a glint reflected from the iris and an arranged shape of the light-emitting diodes are analyzed, whereupon a clear iris image can be obtained. Operation of the iris image obtaining system of FIG. 13 is described below with reference to FIG. 16. FIG. 16 is a flow chart illustrating operation of the iris image obtaining system of FIG. 13.
First, when a user looks at the iris matching point 48 of the translucent mirror 40, it is detected whether an iris image exists in a photographed image or not (step 162). When the iris image exists, the iris images are photographed in real-time during a forward movement of the user to focus the camera 40 on the iris of the user (step S164).
A location of a glint reflected from the iris image is detected by calculating a gray level of the obtained iris image (step SI 66). Thereafter, it is determined whether a dark region exists in the glint region or not (step S168). An image of the lighting means formed on the dark region of the iris image is detected (step S170). And, it is discriminated whether the light-emitting diodes of the lighting mean image are arranged at a regular interval or not (step S172). When the light-emitting diodes of the lighting mean image are not arranged at a regular interval and are vague, the process comes back to the step S168. When the light-emitting diodes of the lighting mean image are arranged at a regular interval and are clear, the obtained iris image is discriminated as a clear iris image (step S174).
FIG. 17 is a block diagram illustrating an entrance door system according to the present invention. The entrance door system of FIG. 17 employs the eye image obtaining method and the iris recognition method described above. FIG. 18 shows an iris image obtaining portion of the entrance door system of FIG. 17. FIG. 19 shows an appearance of the entrance door system of FIG. 17.
The entrance door system of FIG. 17 includes a sensor portion 54, a key pad portion 56, a call portion 58, a voice/text processing portion 60, a memory portion 67, a buffer portion 69, a digital processing portion 70, a comparator 72, a controller 74, a video camera 76, a video phone 78, an entrance door actuator 80, and an iris image obtaining portion 90.
The sensor portion 54 detects an approach of a user and outputs a detecting signal to the controller 74. The key pad portion 56 includes key buttons and is used to input a digit or a letter. For example, a password or an identification (ID) number can be input through the key pad portion 56. The call portion 58 includes a call button (not shown) and is used to a call a person (e.g., staff member of a company) inside an entrance door.
The voice/text processing portion 60 includes a voice processing portion 62, a voice output portion 64, a memory portion 65, and a display portion 66. The voice/text processing portion 60 outputs a corresponding voice message or a corresponding text message in response to a control signal output from the controller 74 through the voice output portion 64 or the display portion 66. The memory portion 65 stores various voice/text messages.
The memory portion 67 stores various data required for an iris recognition such as registered iris images, a user information, and a password. The digital processing portion 70 converts the input iris image into a digital signal and stores it in the buffer portion 69 temporarily. The comparator 72 compares the input iris image with the registered iris image stored in the memory portion 67 and outputs a comparing signal to the controller 74 to determine whether a user is a person allowed to enter the entrance door or not. The controller 74 controls all components of the entrance door system of FIG. 13.
In particular, the controller 74 controls the iris image obtaining portion 90 to obtain a clear iris image. In greater detail, the controller 74 calculates a gray level of the input iris image, and detects a location of a glint reflected from the iris image. Then, the controller 74 determines whether a dark region exists in the glint region or not, and detects an image of the lighting means formed on the dark region of the iris image to determine whether the light-emitting diodes of the image are arranged at a regular interval or not. When the light- emitting diodes of the image are arranged at a regular interval and are clear, the controller 74 regards the input iris image as a clear iris image.
Meanwhile, when a user pushes the call button of the call portion to call a person inside the entrance door, the person inside the entrance door can monitor the user through the video camera 76 and the video phone 78.
The entrance door actuator 80 opens or closes the entrance door in response to a control signal output from the controller 74.
The iris image obtaining portion 90 of FIG. 18 has the same configuration as the eye image obtaining system of FIG. 13. Like reference numerals of FIGs. 13 and 18 denote like parts, and thus their description is omitted to avoid a redundancy. That is, the iris image obtaining portion 90 obtains an iris image by the method of FIG. 16.
Operation of the entrance door system of FIG. 17 is described below with reference to FIG. 20. FIG. 20 is a flow chart illustrating operation of the entrance door system of FIG. 17.
When a user approaches the entrance door, the sensor portion 54 detects an approach of the user and outputs a detecting signal to the controller 74 (step S182). The controller 74 outputs a corresponding voice message or a conesponding text message stored in the memory portion 65 through the voice/text processing portion 60 in response to the detecting signal (step SI 84).
Thereafter, it is determined whether the user pushes the call button of the call portion 58 in a predetermined time period or not (step SI86). When the user is a guest and thus pushes the call button of the call portion 58, the controller 74 outputs a call-related voice message or a call-related text message before the entrance door system is initialized (step S188).
When the user does not the call button of the call portion 58 in a predetermined time period, the controller 74 outputs a voice message or a text message that asks to input an iris of the user. Hence, the user looks at the iris matching point 42 of the translucent mirror 40 to match the iris with a focal axis of the fixed focus camera 44, and then moves forward to focus the camera 44 on the iris. During a forward movement to focus the camera 44 on the iris, the fixed focus camera 44 photographs the iris images in real-time (step S 190).
The photographed iris images are converted into the digital signals by the digital processing portion 70 and then are temporarily stored in the buffer portion 69. The controller 74 calculates a gray level of the input iris image, and detects a location of a glint reflected from the iris image (step S192). Then, the controller 74 determines whether a dark region exists in the glint or not (step S194). When a dark region exists in the glint, the controller 74 also detects an image of the lighting means formed on the dark region of the iris image to determine whether the light-emitting diodes are arranged at a regular interval or not (step S196). When the light-emitting diodes are ananged at a regular interval and are clear, the controller 74 regards the input iris image as a clear iris image (step S 198).
When the clear iris image is obtained, the controller 74 outputs a voice message or a text message through the voice output portion 64 or the display portion 66 in order to notify the user that the input iris image is recognized (step S200). The comparator 72 determines whether the registered iris image stored in the memory portion 67 is identical to the obtained iris image or not (step S202). When the registered iris image stored in the memory portion 67 is identical to the obtained iris image, the controller 74 outputs a control signal to the entrance door actuator 80 to open the entrance door (step S204). The controller 74 determines whether a predetermined time period is passed or not (step S206). After a predetermined time period is passed, the controller outputs a control signal to the entrance door actuator 80 to close the entrance door (step S208).
Industrial Applicability
As described herein before, using the iris image obtaining method according to the present invention, the obtaining process is simplified, whereby increasing a convenience of a user. Also, using the iris image obtaining method according to the present invention, a clear iris image can be obtained. Further, using the iris recognition method according to the present invention, a high quality iris recognition can be achieved. Furthermore, when the iris recognition system using the iris image obtaining method and the iris recognition method is employed in various industrial filed, for example, a security system and a medical field, a high security system an accurate diagnosis can be achieved.
While the invention has been particularly shown and described with reference to preferred embodiments thereof, it will be understood by those skilled in the art that the foregoing and other changes in form and details may be made therein without departing from the spirit and scope of the invention.

Claims

What is claimed is:
1. A method of obtaining an eye image, comprising: a) initializing a photographing unit; b) setting a reference image having a predetermined size in a central portion of an image input through a lens of the photographing unit; c) setting an allowable range of the reference image; d) matching an eye of a user with a focus axis of the photographing unit when the image input through the photographing unit deviates from the allowable range of the reference image; e) photographing the image input through the photographing unit; and f) detecting the eye image among the photographed images; and g) determining whether the eye image is a focused image.
2. The method of claim 1 , wherein the step (b) includes calculating a gray level of each of pixels forming the reference image and a dispersion value of the reference by using a histogram dispersion.
3. The method of claim 1, wherein the allowable range of the reference image is set by a maximum value and a minimum value of the dispersion value.
4. The method of claim 1, wherein the step (g) is performed by using a definition of a glint reflected from the eye image and an arrangement of an image of a lighting unit in a glint.
5. The method of claim 1, wherein the step (f) is performed by a using matching degree between a pupil of the eye image and a circular template.
6. A method of recognizing an iris, comprising: a) detecting a substantial pupil boundary by analyzing a histogram dispersion of an obtained iris image; b) forming a pupil replica boundary; c) calculating a distortion of a pupil by a difference between the substantial pupil boundary and the replica boundary; d) detecting an iris boundary; e) detecting an iris pattern comprised of the pupil boundary and the iris boundary; and
0 determining whether the obtained iris image is a left iris or a right iris.
7. The method of claim 6, wherein the step (a) is performed by a method of binarizing the iris image.
8. The method of claim 7, wherein the step (a) includes detecting a first maximum point and a first minimum point when the histogram dispersion is observed from a dark region to a bright region; and separating a pupil region from an iris region by setting a brightness value at the first minimum point as a critical value.
9. The method of claim 6, wherein the replica boundary is calculated by calculating a central point and a radius of the pupil replica.
10. The method of claim 6, wherein the step (f) is performed by analyzing a location of the central point of the pupil with respect to a location of the central point of the iris.
11. The method of claim 6, further comprising, determining whether a user wears glasses, wherein the histogram dispersion becomes narrower when the user wears glasses.
12. An eye image obtaining system, comprising: a translucent mirror arranged on a front surface of the eye image obtaining system and including an iris matching point; a camera arranged at a location corresponding to the iris matching point to photograph an iris image of a user; a lighting unit for throwing a light on a focal portion of the camera; and an indication lamp indicating that the camera is focused on the iris image.
13. The system of claim 12, wherein the lighting unit includes a plurality of infrared light-emitting diodes arranged at a regular interval, and the lighting unit is tilted by a predetermined degree to emit an infrared ray toward the iris of the user.
14. The system of claim 12, wherein a clear iris image is obtained by analyzing a definition of a glint reflected from the iris and an arranged shape of the light-emitting diodes.
15. The system of claim 12, wherein the camera is a fixed focus camera, and the camera is focused on the iris image such that the user looks at the iris mathching point and moves forward.
16. An entrance door system, comprising: a sensor portion detecting an approach of a person to generate a detecting signal; an iris image obtaining portion for obtaining an iris image; a memory portion storing registered iris images; a comparator comparing the obtained iris image with the registered iris image to a comparing signal; a controller driving the entrance door system in response to the detecting signal of the sensor portion and outputting a control signal in response to the comparing signal of the comparator; and an entrance door actuator opening or closing an entrance door in response to the control signal of the controller.
17. The system of claim 16, further comprising, a digital processing portion converting the iris image obtained by the iris image obtaining portion into a digital signal; a key pad portion including key buttons to input a digit or a letter; a voice/text processing portion outputting a corresponding voice message or a corresponding text message in response to the control signal output from the controller; and a call portion including a call button to a call a person inside the entrance door.
18. The system of claim 16, wherein an iris recognition is performed by calculating a gray level of the obtained iris image to detect a location of a glint reflected from the iris image, determining whether a dark region exists in the glint region or not, and detecting an image of the lighting means formed on the dark region of the iris image to determine whether the light-emitting diodes of the image are arranged at a regular interval or not.
19. The system of claim 16, wherein the iris eye image obtaining portion includes: a translucent mirror arranged on a front surface of the iris image obtaining portion and including an iris matching point; a camera ananged at a location conesponding to the iris matching point to photograph an iris image of a user; a lighting unit for throwing a light on a focal portion of the camera; and an indication lamp indicating that the camera is focused on the iris image.
20. The system of claim 19, wherein the lighting unit includes a plurality of infrared light-emitting diodes arranged at a regular interval, and the lighting unit is tilted by a predetermined degree to emit an infrared ray toward the iris of the user.
21. The system of claim 19, wherein a clear iris image is obtained by analyzing a definition of a glint reflected from the iris and an arranged shape of the light-emitting diodes.
22. The system of claim 19, wherein the camera is a fixed focus camera, and the camera is focused on the iris image such that the user looks at the iris matching point and moves forward.
PCT/KR2001/001797 2000-10-24 2001-10-24 Eye image obtaining method, iris recognizing method, and system using the same WO2002035452A1 (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
AU2002211048A AU2002211048A1 (en) 2000-10-24 2001-10-24 Eye image obtaining method, iris recognizing method, and system using the same

Applications Claiming Priority (8)

Application Number Priority Date Filing Date Title
KR1020000062539A KR100362484B1 (en) 2000-10-24 2000-10-24 The iris image acquisition method and device using fixed focus camera
KR2000/62539 2000-10-24
KR1020000069072A KR100361141B1 (en) 2000-11-20 2000-11-20 Access control method and installation using iris cognition
KR2000/69072 2000-11-20
KR1020000070962A KR20010016241A (en) 2000-11-27 2000-11-27 Automatic eye image acquisition method
KR2000/70964 2000-11-27
KR2000/70962 2000-11-27
KR10-2000-0070964A KR100376415B1 (en) 2000-11-27 2000-11-27 Pupil acquisition method using eye image

Publications (1)

Publication Number Publication Date
WO2002035452A1 true WO2002035452A1 (en) 2002-05-02

Family

ID=27483482

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/KR2001/001797 WO2002035452A1 (en) 2000-10-24 2001-10-24 Eye image obtaining method, iris recognizing method, and system using the same

Country Status (2)

Country Link
AU (1) AU2002211048A1 (en)
WO (1) WO2002035452A1 (en)

Cited By (10)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
FR2881546A1 (en) * 2005-01-31 2006-08-04 Sagem Eye`s reference axis determining method for securing e.g. building, involves capturing and analyzing geometry of eye, and determining reference axis of eye with respect to reference axis of sensor from analyzed geometry
WO2007096605A1 (en) * 2006-02-21 2007-08-30 Xvista Limited Processing an image of an eye
WO2009024981A3 (en) * 2007-08-21 2010-01-07 Visionix Ltd. Multifunctional ophthalmic measurement system
US20120062720A1 (en) * 2008-12-16 2012-03-15 Hyeong In Choi Apparatus and method for acquiring high quality eye images for iris recognition
US20140022371A1 (en) * 2012-07-20 2014-01-23 Pixart Imaging Inc. Pupil detection device
US20140111632A1 (en) * 2012-10-23 2014-04-24 Pixart Imaging Inc. Pupil detection device
CN104253944A (en) * 2014-09-11 2014-12-31 陈飞 Sight connection-based voice command issuing device and method
CN106791353A (en) * 2015-12-16 2017-05-31 深圳市汇顶科技股份有限公司 The methods, devices and systems of auto-focusing
US9854159B2 (en) 2012-07-20 2017-12-26 Pixart Imaging Inc. Image system with eye protection
US10380442B2 (en) 2016-12-21 2019-08-13 3E Co., Ltd. Iris verification method

Citations (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US5291560A (en) * 1991-07-15 1994-03-01 Iri Scan Incorporated Biometric personal identification system based on iris analysis
KR20000050494A (en) * 1999-01-11 2000-08-05 구자홍 Iris recognition system

Patent Citations (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US5291560A (en) * 1991-07-15 1994-03-01 Iri Scan Incorporated Biometric personal identification system based on iris analysis
KR20000050494A (en) * 1999-01-11 2000-08-05 구자홍 Iris recognition system

Cited By (16)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
WO2006082291A1 (en) * 2005-01-31 2006-08-10 Sagem Defense Securite S.A. Method for determining an eye reference axis
FR2881546A1 (en) * 2005-01-31 2006-08-04 Sagem Eye`s reference axis determining method for securing e.g. building, involves capturing and analyzing geometry of eye, and determining reference axis of eye with respect to reference axis of sensor from analyzed geometry
WO2007096605A1 (en) * 2006-02-21 2007-08-30 Xvista Limited Processing an image of an eye
US9192296B2 (en) 2007-08-21 2015-11-24 Visionix Ltd. Multifunctional ophthalmic measurement system
WO2009024981A3 (en) * 2007-08-21 2010-01-07 Visionix Ltd. Multifunctional ophthalmic measurement system
US20120062720A1 (en) * 2008-12-16 2012-03-15 Hyeong In Choi Apparatus and method for acquiring high quality eye images for iris recognition
US20140022371A1 (en) * 2012-07-20 2014-01-23 Pixart Imaging Inc. Pupil detection device
US9854159B2 (en) 2012-07-20 2017-12-26 Pixart Imaging Inc. Image system with eye protection
US10574878B2 (en) 2012-07-20 2020-02-25 Pixart Imaging Inc. Electronic system with eye protection
US20140111632A1 (en) * 2012-10-23 2014-04-24 Pixart Imaging Inc. Pupil detection device
US9530051B2 (en) * 2012-10-23 2016-12-27 Pixart Imaging Inc. Pupil detection device
US10198631B2 (en) 2012-10-23 2019-02-05 Pixart Imaging Inc. Pupil detection device
CN104253944A (en) * 2014-09-11 2014-12-31 陈飞 Sight connection-based voice command issuing device and method
CN106791353A (en) * 2015-12-16 2017-05-31 深圳市汇顶科技股份有限公司 The methods, devices and systems of auto-focusing
CN106791353B (en) * 2015-12-16 2019-06-14 深圳市汇顶科技股份有限公司 The methods, devices and systems of auto-focusing
US10380442B2 (en) 2016-12-21 2019-08-13 3E Co., Ltd. Iris verification method

Also Published As

Publication number Publication date
AU2002211048A1 (en) 2002-05-06

Similar Documents

Publication Publication Date Title
JP2522859B2 (en) Eye position detection device
EP0989517B1 (en) Determining the position of eyes through detection of flashlight reflection and correcting defects in a captured frame
KR101014325B1 (en) Face recognition system and method using the infrared rays
GB2343945A (en) Photographing or recognising a face
JPH0424503A (en) Apparatus for detecting position of eye
JP5271742B2 (en) Sunglass wear detection device
JP2002352235A (en) Apparatus and method for adjusting focus position in iris recognition system
KR20050085583A (en) Expression invariant face recognition
JP2000083930A (en) Personal identifying device using iris
WO2006006299A1 (en) Pupil detector and iris identification device
US11375133B2 (en) Automatic exposure module for an image acquisition system
WO2002035452A1 (en) Eye image obtaining method, iris recognizing method, and system using the same
JP2000331159A (en) Facial image processor
JPH0632154A (en) Driver's condition detector
US11882354B2 (en) System for acquisiting iris image for enlarging iris acquisition range
KR101122513B1 (en) Assuming system of eyeball position using 3-dimension position information and assuming method of eyeball position
JPH04174309A (en) Driver's eye position detecting apparatus and condition detecting apparatus
JP3813043B2 (en) Glasses reflection detection apparatus and glasses reflection detection method
JPH10268372A (en) Photographing device
JPH05297152A (en) Object recognizing device
KR20010016241A (en) Automatic eye image acquisition method
JPH06323832A (en) Interface for vehicle
JP2677010B2 (en) Eye position detection device
CN109241935A (en) A kind of image capture device controlled based on single-sensor and spectrum
KR100880466B1 (en) Method for pupil detection using of red eye effect for iris recognition

Legal Events

Date Code Title Description
DFPE Request for preliminary examination filed prior to expiration of 19th month from priority date (pct application filed before 20040101)
121 Ep: the epo has been informed by wipo that ep was designated in this application
REG Reference to national code

Ref country code: DE

Ref legal event code: 8642

32PN Ep: public notification in the ep bulletin as address of the adressee cannot be established

Free format text: NOTING OF LOSS OF RIGHTS PURSUANT TO RULE 69(1) EPC

122 Ep: pct application non-entry in european phase
NENP Non-entry into the national phase

Ref country code: JP