KR20000035840A - Apparatus for the iris acquiring images - Google Patents

Abstract

PURPOSE: An apparatus for illuminating the eye to obtain an image of the iris is provided which is useful for verifying the identity of users of automated teller machines. CONSTITUTION: At least one wide field of view camera (3) and with an associated illuminator (6,10) obtains sufficient images of a person to be identified so that x, y, z coordinates can be established for the expected position of that person's eye. The coordinates are used to direct a narrow field of view camera (16) and associated illuminators (21,22,23) to take an image of the eye that can be used for to identifying the person using iris verification and recognition algorithms. These illuminators are positioned and illuminated to eliminate or minimize specularities and reflections that obscure the iris.

Description

Iris Image Capture Device {APPARATUS FOR THE IRIS ACQUIRING IMAGES}

Several methods have been proposed as biometrics to detect or identify individuals. These methods include signature analysis, fingerprint capture and analysis, imaging and analysis of eye retinal vascular patterns. Recently, a method of using eye iris having a very fine pattern that is different for each person as a non-contact, non-exposed biometric has been used for more than a few years. This technique is described in US Pat. No. 4,641,349 to Flom et al., US Pat. No. 5,291,560 to Daugman. The system described in these documents requires that a person hold at least one eye in an arbitrary position with respect to the imaging camera taking the iris. Although this process has been applied somewhat satisfactorily, it is not suitable for expedited processing activities such as the use of automated teller machines, unattended output control or automatic execution. Examples include immigration control, sales verification scores, welfare control enforcement, Internet banking, bank loans or account opening, and other financial transactions.

Flom and Daugman's iris appraisal requires a clear, well-targeted image of the eye's iris. Once the image is captured, the contrast of the recognized human's iris with the coded file image is quickly performed. However, conventionally, if a person who has fixed his eye at an arbitrary position does not approach the video camera, a clear image of the iris of the person whose optical system is recognized cannot be obtained quickly and sufficiently. The system is required to quickly obtain a clear picture of the iris of a human or animal at any position away from the optical system. This system is very useful for identifying users of automatic teller machines, as well as other applications requiring personal navigational access to restricted areas or facilities or user awareness. The system is also used to identify patients, suspected criminals and others who cannot be identified in other ways.

Automatic teller machines (ATMs) are widely used for banking transactions. The user wants to insert their identification card, fill in their identification number, and verify their identity relatively quickly. However, anyone who knows the identification number associated with a card can use it. If the thief sees the owner's card use, or finds the identification number on the card or something, and finds the identification number, he can easily withdraw money from the owner's account. Therefore, banks have been studying other reliable methods of verifying the identity of ATM users.

Since the iris recognition method of Flom et al. Has proven to be very reliable, the use of iris recognition has been proposed to verify the identity of ATM users and the application of other remote user recognition or identification. However, for commercially useful applications, there must be a fast, reliable and non-exposed method of obtaining an iris image with sufficient resolution by having an ATM user standing in front of the teller and identifying and recognizing it. In order to allow the user to position the head at a distance from the camera, it is not practical, such as without the use of eyepieces or other fixtures or attachments. Therefore, there is a need for a system that quickly locates the iris of an ATM user and obtains a large amount of images of the iris that can be used for identification and verification. This system is suitable for use with or without an input / output card. The system should also be able to obtain images from a user wearing glasses, contact lenses, ski masks, or other barriers.

The present invention relates to an apparatus and method for illuminating the eye for capturing an iris image.

1 is a front view of a preferred embodiment of the iris image capture device of the present invention.

2 is a side view of FIG. 1;

3 is a plan view of a first embodiment of a narrow field illuminator;

4 is a side view showing an illumination area of the illuminator according to the broken line of FIG.

5 is a side view of a second embodiment of a narrow field illuminator similar to FIG. 4;

6 is a plan view of the first fixture bracket.

7 is a side view showing the position of the illuminator of the illuminator bracket according to the broken line in FIG.

8 is a plan view of the second fixture bracket.

9 is a side view showing the position of the illuminator of the illuminator bracket according to the broken line in FIG.

10 is a block diagram illustrating a control structure of the embodiment of FIG. 1.

FIG. 11 is a front view illustrating reflection of a wide field illuminator appearing in a human eye and glasses. FIG.

12 is a front view illustrating reflection of a narrow field illuminator appearing in a human eye and glasses.

The present invention provides a method or apparatus for obtaining a clear image of an iris of a person whose head is positioned in front of an optical system that receives light reflected from the iris. The system includes one or more cameras and one or more illuminators with or without ambient lighting. It also provides a pan / tilt reflector, or gimbal device and one or more lenses. Light reflected from an object is captured by a gimbal camera or mirror, or directed through a lens to the camera. In a preferred embodiment, a narrow field of view (NFOV) camera is reflected from the pan / tilt reflector to receive light through the lens or directly through the gimbal mounting camera. A second camera and a third camera are provided to obtain a wide field of view (WFOV) image of the object. In some cases, a WFOV camera may be unnecessary if the user knows that they are always within the field of view of the NFOV camera or can be located with the movement of the NFOV camera. The image from the WFOV camera determines the specific location coordinates of the major war, such as the head, shoulder and iris of the person being recognized. Based on the analysis of such images, the pan / tilt reflector or gimbal is adjusted to receive light from the iris or other critical area and directs the reflected light to the narrow field camera. The camera produces enough images to allow iris recognition.

Preferred embodiments include a wide field illuminator that illuminates the face of the recognized person. The illuminator includes a wide field of view camera or a plurality of infrared light emitting diodes positioned around the cameras.

It also provides two or more narrow field illuminators arranged in an array of light emitting diodes. These arrangements are rotatably installed about the horizontal axis and the vertical axis. By using at least two arrangements, specular reflection and reflection can be compensated from glasses, contact lenses, or other objects blocking the iris.

In addition, the arrangement is such that the first light emitting diode has a centerline perpendicular to the base, the second light emitting diode has a centerline perpendicular to the base, and the third light emitting diode has a centerline obtuse with respect to the base. This provides wide field illumination to the array. An additional control system is provided to enable individual illumination of each group of LEDs. The control system allows for selective action of individual diodes. Another alternative is to provide a single light converted to the coordinate system of the image adjusting device.

An image processor is provided for analyzing the image from the wide viewing angle camera, thereby detailing the location of the key point or area on the object or person being recognized. A preferred technique for identifying a user's location is stereoscopic image analysis. Optionally, visible or invisible region imaging or ranging devices such as supersonic, radar, spread spectrum microwaves or thermal imaging or sensing or other optical means may be used.

The system of the present invention is particularly useful for verifying the identity of an automated teller user. The system can be combined with most conventional automated teller machines and many other financial transactions. Image capture and recognition can generally be performed in less than 5 seconds, or most are less than 2 seconds.

Other configurations of cameras, such as one NFOV camera, one WFOV and one NFOV camera, two NFOV cameras, multiple NFOV cameras, and multiple WFOV cameras, may also be applied to other specific purposes, such as many or insufficient constrained or location scenarios. For example, it is used in iris imaging of a telephone booth or using a wireless handset, multi-iris imaging of a crowd, iris imaging of a person, iris imaging of a horse race, or the number of points of sale in a mobile or stationary vehicle.

1 and 2, a device according to a preferred embodiment of the invention is in a housing 1 of 14.5 inches wide, 15 inches high and 18 inches deep. Housings of this size may be placed in or near an automated teller machine or other limited I / O or input location. When coupled or located within the housing of the automatic teller machine, the unit is positioned behind the slope of the photoelectric. In particular, smoke on the slope, or other visually opaque glass, or comparable plastic, obstructs the field of view of the device. The device is positioned to meet the eyes of most users. 2 shows that a human head is recognized in front of the device

In the preferred embodiment shown in Figs. 1 and 2, two wide field-of-view cameras WFOV are provided with lenses 2 and 4, respectively. The rotation and position of the lenses 2, 4 and other components may be varied to adjust to the appropriate space or other application. The wide field illuminator 6 surrounds the lens. Hoods 5 and 7 are provided around the lens 2 and 4 to avoid direct light from the illuminator 6 passing into the camera. The wide field illuminator consists of a set of light emitting diodes 10. For ease of configuration, this diode set is installed on the small circuit board 12. This board is installed in a housing 13 which surrounds the lenses 2 and 4. As shown in Figure 2, a sufficient number of LEDs, including circuit board 12, is provided and positioned to illuminate the head of a person in front of the device. Thus, the wide field illuminator provides an illumination range that includes an area of about 2 feet radius at about 1 foot from the illuminator. This illumination range is shown in FIG. 2 by a solid line extending from the wide field illuminator 6.

Some of the wide field illuminators are located around the WFOV camera to provide near on-axis illumination. On-axis illumination, near-axis illumination, and oblique illumination are imaged with the least amount of shadows that can produce poor surface contours or other artifacts. These eye lights produce shadowless images with good surface appearance. This type of light brightens people so that the iris is easily located, even if this shape becomes unnecessary and impossible. Any shadows from other light sources and camera angles are minimized or eliminated. This type of illumination control can be used for stimulating parasympathetic autonomic nervous system reflections that cause eye blinking or human changes or other reactions. These changes can be used to determine object detection that can be scaled down to establish specific life indicators or to improve imaging resolution.

Light from the wide field illuminator is reflected from the user's face to the lens of the wide field camera lens (2) (4). This allows the WFOV to produce an image from x, y and z coordinates that can be determined by one of the user's eyes. WFOV cameras can also be used to provide secure video images to monitor input and output or other behavior. Get an image of your right eye. However, it should not be chosen as left eye or right eye or both eyes. WFOV cameras can use many techniques, such as depth, depth of focus, with or without stereo, to determine x-y position and distance to the object of interest. A stereo process is used to compare at least part of the image from two wide field cameras to determine x, y, z coordinates. It also provides a watch director 9 to assist in limiting eye position and activity by catching the user's attention.

Once the location of the selected eye is known, the iris portion of the eye should be illuminated to illuminate the iris image used for iris identification and detection. An image with about 200 pixels across the iris portion of the image may be obtained to reliably operate the image recognition algorithm used for identification and identification.

The iris recognition algorithm compares the iris shape of the image and the same shape among the stored images. If the predetermined number of shapes compared is correct, identification is provided. In some cases, at least 75% of the matches are required. Therefore, it is important that there are no mirrors, pseudo light reflections or black shadows that block important portions of the image. When the user wears glasses, the above things happen easily. To overcome this problem, it provides at least two separate illuminators that are combined with the NFOV camera. These fixtures are selectively switched to suit the combined combination to enhance this image structure. For example, reflections on the glasses to make eye detection easier, making iris imaging impossible.

In the embodiment shown in FIG. 1, one NFOV camera 16 is located behind the mirror 18. Light generated from one or more NFOV illuminators 21, 22, 23 is reflected from the selected eye towards the optical subsystem 30, which directs the reflected light towards the NFOV camera. It is also possible to provide a sensor 14 for detecting the degree of brightness of the peripheral illuminance of the identification subject. The information sensed from this sensor determines which illuminators 21, 22 and 23 will provide if illumination is needed. Alternatively, for this light sensing, it is also possible to use one or a plurality of cameras themselves.

The optical subsystem 30 of the embodiment shown in FIG. 1 includes a pan / tilt reflector, which is attached to a rod 33 extending from the motor 34. This structure allows the pan / tilt reflector to rotate around the tilt axis corresponding to the centerline passing through the bar 33. The motor 34 is mounted on the arm 35, which arm is axially coupled to the base 36 by a rod 37. The arm 35 may move along a circumference of the fan shaft corresponding to the centerline of the rod 37. Light emitted from the NFOV illuminators 21, 22, and 23 is reflected from the iris to the pan / tilt reflector 32. The position and orientation of the mirror 32 is determined such that the light is directed to the mirror 18 reflecting light to the NFOV camera 16. The lens of the NFOV camera is mounted to be movable to change the focus or zoom, and the opening of the camera is adjustable. The NFOV camera 16 may be mounted on a moving platform so that its orientation can be turned toward the eye. Therefore, the optical subsection portion shown in FIG. 1 may not be necessary. The preferred optical sub system of the present invention has five degrees of freedom, such as a pan axis, a tilt axis, a focal axis, an aperture axis, and a zoom axis. It is also possible to use systems with smaller degrees of freedom. The pan and tilt axes are used to position the pan / tilt reflector 32 so that a narrow area of the image is correctly input over the sensing array of the NFOV camera 16. It adjusts the movement along the focal axis, aperture axis and zoom axis to focus the object being imaged. In some cases, only NFOV cameras are required, in which case the WFOV cameras are optional.

Depending on the design of the resolution, magnification, focus and size of the imager, the distance between the camera and the lens and the distance between the lens and the object to be imaged are determined. The size of the imager is of great importance in determining the distance from the stove to the imager and affects the depth of focus. The NFOV camera 16 can use a solid state or vidicon camera and can detect various sizes of industrially common 1/4, 1/2, 2/3 or 1 inch diameters. Those skilled in the art will appreciate that an array is possible. Introducing a mirror in the optical system changes the light path without affecting the length of the optical path. The use of such mirrors allows the above optical path to overlap with its own, thus reducing all the required physical lengths needed to carry out the optical design. It will be appreciated by those skilled in the art that a gimbal camera can be used to perform the image adjustment.

The illuminator position in the embodiment shown in FIG. 1 is set to provide an illumination area indicated by the dashed line in FIG. The illumination area is oriented and positioned so that light is reflected from the user's eyes toward the pan / tilt reflector 32. To achieve this result, three illuminators 21, 22, 23 are provided at different positions in the housing 1. Based on information obtained from a WFOV camera or other available locator, the illuminator is directed to direct light to an area where the user's eyes are likely to be located. The illuminator may be installed so that its position or direction is fixed, it may be installed to enable position correction manually, or may be installed in a bracket on which a motor is mounted as shown in FIGS. 6, 7, 8, and 9. The illuminator may be attached to a sliding mechanism that moves along one axis.

It is preferable to use a light emitting diode, a light emitting device of infrared light or near infrared light, or a combination of such devices as a light source. In order to ensure uniform illumination, a lens and a diffuser (not shown) may be used. Infrared rays are effective as light sources because infrared rays pass through glasses or sunglass better than light in the visible region. In addition, infrared rays are not visible to the user or are not well visible. An optical filter can be placed in the light path in front of the camera to reduce the light from the surrounding wavelengths that interferes with obtaining the desired image. In order to optimize the performance of each camera, different wavelength priority filters can be used that pass light of different wavelengths. For example, long wavelength IR may be used for imaging a WFOV camera, and short wavelength IR may be used for imaging a NFOV camera. The NFOV camera then stops responding to the illumination for WFOV. This so-called "speed of light" processing can be very effective. If desired, the LED light source can be strobed. Strobing provides the ability to stop motion. In addition, by using a high intensity light source for a short time and exposing the camera, strobing provides the ability to subdue the ambient light, thereby eliminating background ambient light that may cause interference.

Strobing the NFOV and WFOV fixtures, for example at different times, and each camera optimally optimizes the spatial and temporal characteristics of each device for integrating photos across the appropriate or different time slices. It is available.

As shown in Figs. 1, 3, 4, and 5, the NFOV illuminator is composed of light emitting diodes 20 arranged in a 6x6 array on a circuit board 24. Figs. If the light emitting diode is provided perpendicular to the circuit board, the illuminator has an illumination area as indicated by the diameter b in FIG. The inventors have found that the illumination area can be increased by rearranging or rearranging the direction or position of the light emitting diodes having the configuration as shown in FIG. This increased illumination area is represented by the longer diameter a of FIG. 5. The two rows of diodes at the top of the illuminator have an acute angle to the circuit board 24 and the two rows of diodes at the bottom have an obtuse angle. A circular or other type of illuminator may be placed around the NFOV lens. To minimize the use of artefacts or to increase the availability of biologically relevant properties such as corneal curvature, eye separation, iris diameter, sclera vasculature, etc. Other optical devices, such as dielectric anisotropy analyzers, can be used.

6 and 7 show the brackets used to attach the NFOV illuminators 21, 22, 23 to the housing. This bracket 30 has a U-shaped base 31 having a hole 32, and the bracket is attached to the housing by a screw inserted into the hole 32. A pair of gripper arms 33 and 34 to which the pin 42 is attached is axially coupled to one of the uprights of the base. The same pair of other gripper arms 33 and 34 are also connected to the opposite uprights by a collar 37. The illuminator represented by dashed line in FIG. 7 is fixed between the gripper arms described above by screws or pins 38. Although not shown in the figures, a series of holes are formed along the uprights described above so that the gripper arms can be selectively positioned in one of several positions. In order to prevent rotation of the link ring, the locking tab 39 extends from the link ring 37 into an adjacent slot of the upright portion. In order to prevent the gripper arm from rotating, the mounting screw 38 is tightened with respect to the pin 42 extending from the gripper arms 35 and 36.

8 and 9 show a second bracket 44 used to attach the NFOV illuminators 21, 22, 23 to the housing 1. The bracket 50 has a base 51 attached to the housing described above. The rod 52 extends upwardly from the base 50. The connecting ring 53 is installed to slide along the rod 52 and may be fixed at a desired position by the mounting screw 54. The rod 55 extends from the connecting ring 53 and holds the carrying section 46. This carrier is attached to the rod 55 in the same manner as the linkage 53. An illuminator shown in dashed line in FIG. 9 is attached to the carrying section 56 by a stator or a snap fixed to the pin 57 extending from the carrying section. With this bracket 50 and with the bracket 30 described otherwise, the fixtures attached thereto are repositioned or repositioned along the pan axis or tilt axis.

Each array of light emitting diodes provided via the distribution substrate 60 is preferably connected to the lighting controller 62 shown in FIG. There may be more than one illuminator, such an arrangement of light emitting diodes is shown in the figure from illuminator 1 to illuminator X shown in the figure. The WFOV illuminator 6 and the NFOV illuminators 21, 22, and 23 in the embodiment of FIG. 1 selectively emit light by the distribution substrate 60 and the illumination controller 62. FIG. In addition, a series of light emitting diodes in each array can be selectively emitted for each array, and individual light emitting diodes can also be selectively emitted.

Each array of light emitting diodes can emit light of a different wavelength for each array. Note that the iris responds differently to light, responding better at some wavelengths and worse at other wavelengths. In order to solve this problem, various illuminators having light emitting diodes having different wavelengths may be used, or one illuminator having light emitting diodes having different wavelengths may be used. The light emitting diodes can be strobed, whereby a better image can be obtained. In addition, it is also possible to install light emitting diodes of different light widths in parallel with other illuminators in order to control the brightness of the illumination or to control the specularity. At this time, the denser the light width, the smaller the magnitude of the specular reflectance.

In addition, it is possible to control the light emission time and the brightness. In order to prevent the illuminator from turning off due to prolonged illumination, a timer 63 may be installed in each illuminator, in which the timer 63 is represented by a dashed block labeled “T” in FIG. 10. The timer stops powering the fixtures in the row after a period of illumination. The WFOV camera 3 provides an image to an image processor 64 called PV-I. The controller 64 provides the computer 65 with x, y, z coordinates with information about the position of the eye of the identity subject. The image processor may evaluate the image quality and may include an algorithm for compensating for the movement of the subject. The image processor can also improve the quality of the image. The computer 65 evaluates the information from the ambient light level detector 69 and the NFOV camera. This data is used to modify the lighting method. Using the x, y, z coordinates for the expected position of the eye, the computer can control the direction of the lighting controller that controls the lighting of the illuminator, and the pan / tilt controller The orientation can be controlled, thereby obtaining useful image information of the iris. Depending on the processing result of the WFOV camera, these functions can be selected. The computer 65 sends commands to the motor to correct or focus the position of the pan / tilt axis. In the simplest case, a WFOV image may be obtained, and after the data has been processed, it may be considered that the data is passed to the pan / tilt controller 66 or horizontal controller via the image processor 64 and the computer 65. To minimize run time and control adjustment time, the motion along all five axes in the optical subsystem can be synchronized.

The pan / tilt controller 66 receives a macrolabel command from a computer and generates correct setpoints and / or commands for control of illumination or monitoring of each axis. A set of consecutive intermediate paths for the axis takes place here and is then sent to the supervisory controller for each axis. The command interpreter interprets the command from the image analysis and formats the response using the location information from the optics. A real time interrupt generates a fixed clock signal every nms. This signal is necessary for the performance of the sampled data system for the position controller of each axis, and the monitoring controller enables synchronization for the movement of the continuous path. The diagnostic subsystem diagnoses whether the state of the control system is normal.

In addition to the five axis choreography described above, these microprocessors control illumination. In order to operate the selected illuminator at a suitable time or to operate the selected illuminator simultaneously with taking a camera frame, the lighting controller is provided with a command such as a command related to motion control.

The image from the WFOV is transmitted to the image processor 64 as an analog signal. Two pyramid processors, memory for storing at least two frames, one LUT, an ALU device, a digitizer to digitize analog signals, a Texas Instruments TMS 320 C-31 or C-32 processor, Preferably, the image processor includes a serial / parallel processor. The above image is processed using a pyramid processor such as that described in US Pat. No. 5,359,574 to van der Wal. The Texas instrument processor calculates whether there is a mismatch between the images. The WFOV image defines an area or point corresponding to the position of the right, left, or both eyes of the subject within the field of view of the WFOV camera. When there is such a mismatch, using stereo processing techniques, the point representing the subject for the WFOV camera is represented by x, y, z coordinates. This information is further processed to define areas of interest such as heads or eyes. The coordinates of the region of interest are used to set the orientation of the NFOV optical system. These position coordinates are transferred from the image processor to the NFOV image and the iris image processor 65. The device 65 includes a 486, Pentium, or other microprocessor system and associated memory. The memory stores programs or algorithms for orienting the optical platform and recognizing the iris. The WFOV video image can also be stored as a secure video record.

The focal axis of the NFOV system can be controlled in an open loop manner. In this case, the x, y, z coordinates obtained by stereo processing define the focal axis position by analytical computation or table look, so that the lens focus of the NFOV camera can be correctly aligned with the desired object. . Closed loop focusing can also be used. In this case, the NFOV video may be processed by the image processor 64 to obtain a figure of merit that determines whether the axis is in focus. A new image is obtained by receiving a command to move the axis back and forth from the sensitivity index. This process proceeds in a closed loop until the image is in focus. Other information, such as the size and position of the iris measured by the camera device and eye saparation obtained from the WFOV camera image, are combined with stereo and other focus information to become multivariate features. This multivariate feature is used to make range information more accurate by fusing direct or derived sensor information. Of course, such sensor fusion may include other information.

Since the object of interest, i.e., the eye, is moving, it is necessary to track the trace detected by the WFOV by the NFOV camera. When the image stops being obscured by the operation, an image of good quality can be obtained by the NFOV camera and the optics. When you want to obtain a good quality iris image, align the above-mentioned optics that direct light to the NFOV camera so that no further action is required.

In this case, at some uniform sample rate, for example, the x, y, z coordinates obtained from the analysis of the WFOV image are sent to the NFOV controller every 100 ms. Continuous path algorithms as described in Robotic Engineering An Integrated Approach (Prentice Hall, 1989) by Clafter, Chmielewski and Negin. Can be used to provide an intermediate set of setpoints {p, t, f, a, z} to the axis, so that the axis continues to move while in the tracking state. To define the position of the last end, macro level instructions may be given, or the same set {p, t, f, a, z} may be sent during the sample period.

As the NFOV axis moves, there may not be enough time for the associated imager to perform the integration required to obtain a clear image. In addition, depending on the camera used for interlaced or sequential scanning or the like, there may be a field to field arrangement or vertical arrangement of the image, all of which may be corrected in whole or in part by calculation. As such, it is easy to see why the WFOV camera provides the necessary information to determine the direction of the NFOV stage. Although the image is slightly obscured by movement focus or exposure blot, eye tracking algorithms such as reflectance, iris structure, or iris conformation provide sufficient information to give the NFOV camera a reasonable judgment of the eye's position. It should be noted that it can provide. Conceptually, WFOV data may be used for coarse movement, and NFOV data processed as above may be used for precise resolution as additional information during operation. In determining the position of the NFOV image to obtain an iris image with good image quality, this fusion of data provides better judgment than using a single WFOV camera image.

In order to obtain an iris image with good image quality, the NFOV axis should be fixed at a point where the residual motion is smaller than can be detected by the imager. If this occurs, any residual image (typically there is a delay between the integrated image and the reading by RS170) and the appropriate allowance time used for clear images should be removed from the imager. See Robotic Engineering An Integrated Approach to learn about timing scenarios. This can be accomplished by several methods, the simplest of which is the time lag that occurs after stopping the operation until a good quality RS170 image is obtained. Multiple iris images, which may be partially indeterminate, are collected and merged, resulting in a less obvious mixed iris image using optimization and fusion methods.

The inventors have discovered that any light source causes reflections in the glasses of the identity subject. 11 is a view showing eyes detected by the WFOV camera 3. There is reflected light 70 by the WFOV illuminator 6 on the subject's glasses 74. The reflected light 70 partially obscures the iris 76 of the subject's eye, making it difficult to recognize the iris, if not impossible. In order to solve this problem, illuminators 21, 22 and 23 which are in an off-axis position from the optical axis of the NFOV camera 16 are used. The position of the light emitting diodes can be determined deeply, and only a small number of light emitting diodes can be used to provide a sufficient amount of illumination without causing reflections that obscure iris identification. These results are shown in FIG. From this figure, it can be seen that the reflected light 80 of the light emitting diodes of the NFOV illuminators 21, 22, and 23 only appears in the image. The reflected light does not cover any part of the iris 76. In some cases, especially when wearing glasses, the reflection of the WFOV can make head and eyes look faster.

Using multiple illuminators, the shape of the glasses can be determined by the images in the image corresponding to the glasses. The glasses are sequentially illuminated using two illuminators separated from each other. During the first illumination the specularity is in one position and during the second illumination the reflection is in a different position. The appropriate amount of spectacles is determined by calculating the total amount of the reflected image change. From this information, it is possible to determine the minimum movement required to bring the reflected image away from the iris.

A scale correction process is used to correspond to the center portion of the NFOV camera's field of view for the position of the pan / tilt and focal axis for a series of coordinates defined in three-dimensional space by the wide field of view (WFOV). Given the WFOV coordinates that determine the position of the user's eye in the motion space in front of the camera, the coordinate axes of pan, tilt and focus {p, t, f} can be determined by means of a transducer or table lookup. The above-described coordinate axis causes the center of the NFOV camera field of view to be aligned on the x and y axes, and the focus is aligned to the z axis. It is advisable to use a series of target points to aid in calibration. This target point partially fills the circle corresponding to the position of the iris at a given position on a page. The target point is located at a predetermined distance from the housing, and the device is operated to locate the iris and generate a calibration image.

If you use NFOV and WFOV cameras, you need to calibrate both cameras together. Such calibration may be accomplished by manual or automatic processing and may use optical or projection targets. The automatic processing method may require a calibration target to be automatically recognized by the calculation assistant for driving the camera image. It is also possible to project the target in the calibration volume using the NFOV camera's motion function or other motion generation function and then recognize the target.

Although the present invention has been described above by the preferred embodiments of the compact image adjusting and focusing apparatus and the method of using the above apparatus, the present invention is not limited thereto, and various methods within the scope of the claims described below. It is clear that it can be carried out as.

Claims (34)

One or more cameras arranged to take images of the eye such that the images contain representations of the iris with sufficient resolution to be used for iris identification and identification; One or more illuminators arranged for illuminating the iris and composed of a plurality of light emitting elements that can be selectively and simultaneously revealed; Iris image capture device having a. The iris image capture device according to claim 1, further comprising a second illuminator arranged to illuminate the iris, wherein the second illuminator is arranged apart from the first illuminator. The iris image capture device according to claim 2, wherein the second illuminator is composed of a plurality of light emitting elements that can be selectively identified. The iris image capture device according to claim 1, wherein light is reflected from the iris to reach one or more cameras along the camera center line, and the light travels along a path crossing the camera center line from the one or more illuminators. The iris image capture device according to claim 1, wherein the one or more illuminators comprise an array of one or more light emitting diodes disposed on the base. The method of claim 5, wherein the array of light emitting diodes, A first set of light emitting diodes attached to the base in a manner that emits light along the first path; A second set of light emitting diodes attached to the base in a manner that emits light along a second path that is not in parallel with the first path; Iris image capture device comprising a. 6. An iris image capture device according to claim 5, wherein the first set of light emitting diodes and the second set of light emitting diodes can be selectively identified. 6. An iris image capture device according to claim 5, wherein the iris image capture device comprises an array bracket, and wherein the base of the array of one or more light emitting diodes is pivotally attached to the array bracket to rotate about the center of the array. . 9. The iris image according to claim 8, wherein the iris image capturing apparatus includes a second bracket, and an array bracket is attached to the second bracket so as to move the array along a line perpendicular to the center of the array. Capture device. 6. The iris image capture device according to claim 5, wherein at least some of the light emitting diodes emit light having a wavelength different from that of light emitted from other light emitting diodes. The iris image capture of claim 1, wherein the one or more illuminators can emit one or more of infrared, visible light, near-infrared light, selected bands of light frequencies, and can emit both visible and infrared light. Device. The iris image capture device according to claim 1, wherein at least one illuminator emits light of varying intensity. 13. The apparatus of claim 12, comprising a power source connected to one or more fixtures that emit light of varying intensity, and a controller connected to the power source to convert the power output to the one or more fixtures. An iris image capture device, characterized by varying the intensity of light emitted by the illuminator. The iris image capture device of claim 1, wherein the one or more illuminators emit different wavelengths of light. 3. An iris image capture device according to claim 2, characterized in that the iris image capture device comprises a controller coupled to at least one illuminator and a second illuminator, the controller comprising a program for selectively illuminating the illuminators according to a dynamically predetermined pattern. Iris image capture device. The iris image capture device according to claim 1, further comprising a wide field of view camera arranged to take an image including an eye. The iris image capture device according to claim 16, comprising a wide field illuminator. 18. The iris image capture device according to claim 17, wherein the wide field illuminator is composed of a ring of light emitting elements arranged around the wide field camera. 18. The iris image capture device according to claim 17, comprising a hood surrounding a wide field of view camera to prevent light from the illuminator coming directly into the camera. The iris image capture device according to claim 1, further comprising a power source connected to at least one illuminator and a timer connected to the power supply and turning off the power source after a selected time period. 10. The apparatus of claim 1, comprising a roughness sensor and a controller coupled to the ambient light sensor and one or more illuminators, the controller comprising a program to turn off the fixtures when the ambient light is at a predetermined level. Iris image capture device characterized in that. The iris image capturing apparatus of claim 1, further comprising a motor connected to one or more illuminators. The iris image capture device according to claim 1, comprising an optical system for receiving the light reflected from the iris and sending the light to one or more cameras. The iris image capture device of claim 1, wherein the one or more cameras are movable. In the iris image capture method, Positioning a three-dimensional coordinate position of the eye including the illuminated iris; Positioning the camera such that any light reflected from the iris can be reflected to the camera; It consists of a plurality of light emitting elements that can be selectively revealed so that light is reflected from the iris along the camera center to the camera and the light travels from one or more illuminators along one or more paths intersecting the camera center. Illuminating the eye with one or more illuminators; Generating at least one iris image having sufficient resolution with the camera while the illumination is used for identification and identification; Iris image capturing method comprising a. 27. The apparatus of claim 25, wherein the one or more illuminators are comprised of first and second illuminators that are illuminated in series, while the continuous illumination is to identify and identify features of the iris bundle, while the camera is configured to produce images of sufficient resolution. An iris image capturing method comprising generating a first image and a second image to be used. 26. The method of claim 25, wherein the one or more illuminators are comprised of two sets comprising a plurality of light emitting elements that can be selectively revealed and subsequently illuminated. 27. The iris of claim 25, wherein the one or more illuminators can emit at least one of one selected band of infrared light, visible light, light in the vicinity of the infrared light, and light frequencies, and can emit both visible light and infrared light. How to capture images. An iris image capturing method according to claim 25, wherein the image is composed of a plurality of pixels, and a part of the image including the iris is composed of 200 pixels or more. The method of claim 25, wherein the three-dimensional coordinates of the eye, a. Can be determined by using a wide field camera to generate a first image of the area where the eye is believed to be located; b. Can be determined by using a second wide field camera that is spaced from the first wide field camera to produce a second image of the area where the eye is believed to be located; c. An iris image capturing method, which can be determined by synthesizing a first image and a second image to determine a three-dimensional coordinate position of the eye. 31. The method of claim 30, wherein the images are synthesized using stereographic image analysis. 31. The method of claim 30, comprising illuminating the area using near center illumination. 27. The system of claim 25, wherein the camera is disposed within an optical subsystem that includes a pan / tilt reflector such that the reflected light is directed to the camera from the pan / tilt reflector and that the light reflected from the iris is sent to the camera. And adjusting the pan / tilt reflector toward the three-dimensional coordinate position. 27. The method of claim 25, wherein the camera is provided with a gimbal, comprising adjusting the camera toward three-dimensional coordinates of the eye to send light reflected from the iris to the camera.