A device for iris recognition when passing through.

Abstract

The present invention relates to a device (100) for iris recognition during passage, comprising: a first iris camera (10) having a first field of view (11) and a first focal plane (12); a second iris camera (20) having a second field of view (21) and a second focal plane (22); a scene camera (30) having a third field of view (31); and a control unit (50). The first iris camera (10) and the second iris camera (20) are height-adjustable. The first iris camera (10), the second iris camera (20), and the scene camera (30) are connected to the control unit (50). The control unit (50) is configured to use the scene camera (30) to detect an image of the person to be identified (80) located outside the combined detection area of ​​the first iris camera (10) and the second iris camera (20), to determine the height (82) of the periocular facial region (81) of the person to be identified (80) based on the image detected by the scene camera (30), to adjust the heights of the first iris camera (10) and the second iris camera (20) in response to the determined height (82) of the periocular facial region (81), to detect the periocular facial region (81) using the first iris camera (10) and the second iris camera (20), and to confirm the personal identification of the person to be identified (80) based on the detected periocular facial region (81).

Description

Technical Field

[0001] The present invention relates to an apparatus for performing iris recognition of a person when passing through.

Background Art

[0002] Background of the Invention For confirming access rights or identifying other personal identifiers of a person, generally, acquisition of biometric features such as fingerprint, voiceprint recognition, iris recognition, etc. can be performed.

[0003] Generally, in iris recognition, the eyes of a person in an image of the entire eye or the periorbital facial region (i.e., the part of the person's eye) of the person in the form of an individual image are detected, and then the iris features of the person are compared with the stored data. Such an image of the person to be identified can be taken, for example, in a stationary device equipped with a corresponding camera. However, such personal identifier inspection in a stationary device takes time and does not have particular comfort. This is because the person to be identified cannot easily move forward, but first has to stop at the location of the stationary device and have themselves photographed by the camera.

[0004] As another means of iris recognition, it is possible to monitor a detection corridor through which the person to be identified passes using a corresponding camera, and perform detection of the periorbital facial region while the person is passing through the detection corridor. It is clear that this form of iris recognition is much faster and more comfortable because the identification is performed while the person moves through the detection corridor without stopping. However, human behaviors such as blinking, the camera being out of sight, or reflections from glasses make it difficult to detect an appropriate iris image for successful personal identifier identification of the person passing through. For this reason, usually, in this type of device, in order to increase the probability of detecting at least one appropriate image for iris recognition, attempts are made to take as many clear iris images as possible over as large a detection area as possible for the person to be identified.

[0005] In this case, devices of this type that enable iris recognition in transit typically use separate high-performance cameras, such as 3D cameras (e.g., light field cameras (plenooptic cameras)) or cameras with adjustable working distance / focus planes, to enable iris detection at various distances from the camera while a person moves through the detection area. However, such complex systems, on the one hand, limit the possible frame rate, resulting in a small number of sharp images. Furthermore, the resulting images are often susceptible to artifacts that impair image clarity. High image clarity is one of the central requirements for robust iris recognition. [Overview of the Initiative] [Problems that the invention aims to solve]

[0006] explanation Therefore, the object of the present invention is to provide a reliable iris recognition system for passing objects that eliminates the need for a complex camera system. [Means for solving the problem]

[0007] The above-mentioned problems are resolved by the subject matter of the independent claims. Illustrative embodiments will become apparent from each dependent claim and the following description.

[0008] According to one embodiment, a device for iris recognition during passage is provided. The device comprises a first iris camera having a first field of view and a fixed first focal plane, a second iris camera having a second field of view and a fixed second focal plane, a scene camera having a third field of view, and a control unit. The first and second iris cameras are height-adjustable. The first iris camera, the second iris camera, and the scene camera are connected to the control unit. The control unit is configured to detect an image of a person to be identified located outside the combined detection area of ​​the first and second iris cameras using a scene camera, determine the height of the periorbital facial region of the person to be identified based on the image detected by the scene camera, adjust the heights of the first and second iris cameras in response to the determined height of the periorbital facial region so that the first and second iris cameras can detect the periorbital facial region, detect the periorbital facial region using the first and second iris cameras, and confirm the personal identification of the person to be identified based on the detected periorbital facial region.

[0009] The apparatus according to the present invention can, in particular, monitor a detection corridor through which a person to be identified passes (as described below). A detection corridor, in this context, may be, for example, a passageway or walkway that appropriately guides a person to be identified to pass alongside a first iris camera, a second iris camera, and a scene camera. However, generally speaking, a detection corridor is merely a functional term for the detectable area of ​​the camera apparatus used and does not necessarily have to be physically separated (for example, by an appropriate barrier).

[0010] The scene camera, the first iris camera, and the second iris camera are cameras that each have a fixed focal plane, i.e., a fixed working area / working distance from the camera.

[0011] The scene camera specifically monitors the area outside the detection area formed by the first and second iris cameras combined, as described later, and more specifically the entire area on the entrance side of the detection corridor or the area surrounding the entrance side of the detection corridor. This refers specifically to the area outside the detection corridor, as well as the area between the entrance side of the detection corridor and the detection area formed by the first and second iris cameras combined. However, of course, the scene camera can also additionally cover the detection area formed by the first and second iris cameras combined. When the personal identification process is initiated (for example, when a person enters the monitoring corridor, or when such a personal identification process is manually initiated), the scene camera captures an image of the person entering and transmits the image to the control unit.

[0012] The control unit may be any suitable control unit, for example, a general-purpose computer with corresponding memory and processor components, an ASIC, an FPGA, or any other suitable control unit or combination of such control units.

[0013] The control unit extracts the height of the peri-facial region of the person to be identified as they pass through the monitoring area from the scene camera image. In particular, the scene camera may have a focal area around the entrance side of the monitoring corridor, thereby enabling the capture of a clear image of the person to be identified at this location. The scene camera can also basically record images of the person to be monitored at other locations along the monitoring corridor. The clarity of the scene camera image is not an excessive limiting factor, because the scene camera is only responsible for calculating the height of the peri-facial region, that is, enabling control to allow for coarse orientation of the system (particularly the first and second iris cameras). Therefore, the quality of the scene camera image only needs to be good enough to enable such height determination. As can be easily understood here, an excessively clear image is not necessary.

[0014] The scene camera transmits the detected image to the control unit, which uses a corresponding image evaluation algorithm to determine the facial region around the eyes of the approaching person, and thus its height, based on the captured image material.

[0015] The first and second iris cameras themselves are configured to capture the original images used / required for iris recognition. For this purpose, the first and second iris cameras each have a corresponding field of view and a corresponding focal plane. Generally, the field of view of a camera represents the angular range detected by each corresponding camera. Furthermore, the focal plane represents the distance in front of each camera from which the sharpest image is captured. Generally, the focal plane is defined by the focal length of each camera.

[0016] The first and second iris cameras are cameras with a fixed focal plane and preferably a small field of view, so that images can be captured without significant distortion. Note that if the field of view is large, a high number of pixels in the camera's image sensor is required to enable the resolution necessary to extract the corresponding features (e.g., iris features). Accordingly, an iris camera with a small field of view can be used with an image sensor that has a reduced number of pixels. The first and second iris cameras are preferably cameras with a high frame rate, particularly a frame rate of at least 10 images per second (fps).

[0017] Furthermore, the first and second iris cameras are height-adjustable, thereby allowing adjustment of their imaging height. In this case, such height adjustability can be achieved by any conceivable method. For example, the first and second iris cameras can be mounted on separate, height-adjustable holders, or on a common, height-adjustable holder, as described below, and these holders can be driven and controlled by a control unit. It is also possible to mount the iris cameras themselves in a fixed position, allowing only adjustment of the detection area (e.g., by a mechanism or optical system).

[0018] The control unit, after determining the height of the periocular facial region of the person to be identified using the scene camera, adjusts the heights of the first and second iris cameras to match the height of the periocular facial region of the person to be identified, which was previously determined based on the image from the scene camera. This height adjustment capability allows for the use of cameras with particularly small fields of view because the camera is shifted or adjusted to the height of the eyes of the person to be identified (i.e., the height of the periocular facial region), enabling detection of the periocular facial region even with a small field of view.

[0019] By using two iris cameras, each with its own fixed focal plane, it is possible to capture images clearly, or at least clearly enough to extract iris features, at various distances in front of the camera setup (i.e., in front of the first and second iris cameras). This enables iris recognition of the person to be identified by a simple camera with a fixed focal plane, i.e., a fixed focal length or working distance in front of the camera, while the person to be identified moves along the detection corridor. In particular, this configuration avoids complex camera systems such as plenoptic cameras or other refocusable cameras. This allows the first and second iris cameras to operate at higher frame rates, significantly increasing the probability of successful personal identification. Furthermore, by eliminating unnecessary complexity, a more reliable and robust overall system is achieved.

[0020] In summary, the disclosed apparatus uses a scene camera with a wide field of view and, preferably, two or more iris cameras, each having a smaller field of view and a fixed focal plane. The scene camera performs coarse detection of the periorbital facial region or its height of the person to be identified. The height of the iris camera (i.e., its detection height) is then adjusted to match the height of the periorbital facial region, and images of the periorbital facial region are captured while the person moves through a corresponding detection corridor. Image data or characteristics of the user's iris to be identified can then be extracted from these images, and these are also used for personal identification of the person. The disclosed apparatus allows simple cameras with a fixed focal plane to be used as iris cameras. The frame rate is increased, and as a result the images become sharper, the probability of successful personal identification is improved. In this case, the use of multiple cameras allows for personal identification while the person to be identified is moving along the detection corridor.

[0021] The disclosed device enables personal identification for purposes such as access monitoring or any other personal identification triggered therefrom. For example, in access monitoring, a door or gate can be automatically opened if an authorized person is identified. The control unit may also have a display that outputs the detected personal identification to, for example, security personnel. Exemplary applications include personal identification in airports, train stations, courthouses, corporate buildings, etc. Note that this is only an example.

[0022] Although this specification describes the case with two iris cameras, it should be noted that the disclosed apparatus may have two or more iris cameras, for example, three, four, five, or any number of iris cameras.

[0023] In one embodiment, the device for iris recognition during passage further includes a height-adjustable holding section, which supports a first iris camera and a second iris camera and is connected to a control unit. The control unit is configured to set the heights of the first iris camera and the second iris camera by appropriately driving and controlling the height-adjustable holding section.

[0024] The height-adjustable support may be, for example, a linear actuator or other motor or similar device to which the first and second iris cameras are fixed. For example, such a linear actuator may be represented by a motor-driven, height-adjustable holder in a corresponding column to which the iris cameras are fixed. However, this is just one example. Any height-adjustable holder that allows for height changes of the iris cameras can be configured.

[0025] Rather than fixing two iris cameras to a single common height-adjustable holder, each of the iris cameras can be fixed to a unique height-adjustable holder. When two or more iris cameras are provided, all of the iris cameras can be fixed to a common support part that is height-adjustable, or each iris camera can be fixed to a unique support part that is height-adjustable. Further, some of these iris cameras may be fixed to a single common holder respectively.

[0026] According to another embodiment, the apparatus for iris recognition during passage further includes a stroboscopic lamp that operates in the near infrared region during passage. The control unit is configured to drive and control the stroboscopic lamp during the detection of the periorbital facial region using the first iris camera and the second iris camera so that the stroboscopic lamp emits light pulses during the detection of the periorbital facial region.

[0027] When the light pulse enters the eyes of the person to be identified simultaneously with the imaging by the iris camera, the recognition of iris features becomes easy. For safety reasons, the eyes of a person should be irradiated with only a predetermined amount of light per unit time. The use of stroboscopic light enables an increase in light intensity without exceeding the allowable light energy introduced into the eyes in time average.

[0028] Also, during the operation of the stroboscopic lamp, continuous distance measurement is performed (for example, using laser distance measurement, infrared distance measurement or any other appropriate distance measurement), and based on the distance, the intensity of the light entering the eyes can be adjusted so as not to exceed the allowable light intensity. That is, in this case, the light intensity is reduced as the distance decreases.

[0029] According to another embodiment, the light pulse is synchronized with the image acquisition by the first iris camera and the second iris camera.

[0030] By synchronizing the light pulse with image capture, the iris of the person to be identified is illuminated by the iris camera during the capture of each image, thus ensuring that iris features are clearly identifiable and extractable.

[0031] According to other embodiments, the third field of view is larger than the first and second fields of view.

[0032] The scene camera (corresponding to the third field of view) is used for recognizing the periorbital facial region and orienting the iris camera to the height of the iris camera, and therefore must detect a larger angular range. However, a smaller field of view generally results in less image distortion, such as chromatic aberration, excessive spherical distortion, corner blur, and vignetting. Since the scene camera is used only for coarse orientation toward the height of the periorbital facial region, such distortion is not very important here. However, in order to identify iris features, it is desirable that the periorbital facial region and, by extension, the iris of the person to be identified, be captured with as much detail and fidelity as possible. Therefore, the iris camera has a smaller field of view, which allows for more detailed and fidelity capture of the iris and enables the use of an image sensor with a lower pixel count.

[0033] In another embodiment, the control unit is configured to detect the periorbital facial region using a first iris camera and a second iris camera while the person to be identified moves through a detection corridor.

[0034] In other embodiments, the first iris camera has a first depth of field, and the second iris camera has a second depth of field.

[0035] Depth of field is a measure of the extent of the sharp area within the object space of an imaging optical system, such as a camera. In other words, depth of field represents the size of the distance range of a camera in which an object is sufficiently sharply imaged. Since sharp imaging occurs in the focal region or focal plane, a large depth of field means a large area centered on the focal plane that is sharply imaged.

[0036] The first and second iris cameras may both have equivalent depths of field or different depths of field. However, the defined focal length also depends additionally on the position of the focal plane. Therefore, even if the two iris cameras have equivalent depths of field, the depths of field of the first and second iris cameras (described below as detection volume) may deviate from each other. This allows the first and second iris cameras (and other iris cameras) to clearly detect various regions along the detection corridor, thus enabling the common capture of clear images of the periocular and facial regions over a wider area than with an iris camera alone.

[0037] In other embodiments, a first focal plane and a first depth of field define a first detection volume, and a second focal plane and a second depth of field define a second detection volume. The first detection volume and the second detection volume each represent a spatial volume in front of the corresponding detection volumes of the first and second iris cameras, respectively, which are clearly imaged by the corresponding detection volumes of the first and second iris cameras.

[0038] According to other embodiments, the first detection volume and the second detection volume are at least partially different from each other.

[0039] In particular, the first and second detection volumes are positioned sequentially in the direction of the detection corridor, allowing the first and second iris cameras to clearly image various regions along the detection corridor. This enables the capture of clear images of the periorbital and facial regions over a wider area than when using only a single iris camera. Consequently, the success rate of identifying the detected dataset and, consequently, the personal identification rate is improved.

[0040] Since the first and second iris cameras are oriented to the same height (height of the periorbital facial region) during the personal identification process, the detection volumes will be substantially equivalent in the vertical direction, except when the first field of view of the first iris camera and the second field of view of the second iris camera are misaligned. However, the detection volumes of the first and second iris cameras may differ from each other within a limited range, provided that the periorbital facial region is guaranteed to be detectable by both iris cameras.

[0041] The fact that the first detection volume and the second detection volume differ from each other, in particular, means that there may be an overlapping region between the detection volumes (especially along the detection corridor, i.e., along the horizontal distance from the iris camera) that can be clearly imaged by both the first and second iris cameras. This enables continuous and clear image acquisition along both the first and second detection volumes.

[0042] In another embodiment, the first detection volume and the second detection volume are located one behind the other within a detection corridor in which the person to be identified moves during iris recognition.

[0043] The detection corridor extends toward the first and second iris cameras in at least the region of the first and second detection volumes, so that the person to be identified moves toward the first and second iris cameras, at least partially, within the detection corridor.

[0044] In other embodiments, the first iris camera, the second iris camera, and the scene camera are each non-plenooptic cameras.

[0045] As already shown in the background of the present invention, a plenooptic camera (also called a light field camera) that can achieve a higher depth of field is a high-performance camera. However, in this type of camera system, the frame rate decreases due to the large amount of data, and consequently, the number of sharp images produced decreases. Therefore, in the disclosed apparatus, such a camera is omitted, and instead, a number of simple cameras having mutually different fixed focal planes are used. This enables a large and sharp detection area simultaneously with a high frame rate.

[0046] According to another embodiment, the device for iris recognition during passage further includes a recognition device connected to a control unit, which is configured to identify the entry of a person to be identified into a detection corridor and to initiate a personal identification process in response thereto.

[0047] Therefore, the personal identification process can be automatically triggered by the control unit when the recognition device detects that the person to be identified has entered the detection corridor.

[0048] However, it should also be noted that the personal identification process can be initiated manually, for example, by operating the corresponding input device.

[0049] According to other embodiments, the recognition device includes at least one of a photointerrupter, an NFC device, an infrared distance sensor, and a motion sensor mounted on the entrance side of the detection corridor.

[0050] In an infrared distance sensor, for example, the personal identification process can be automatically initiated depending on the distance to the person to be identified.

[0051] A photointerrupter or motion sensor can, for example, identify the moment a person enters or approaches a detection corridor, and the control unit can automatically initiate the personal identification process accordingly.

[0052] An NFC device can perform identification, for example, when a transponder (e.g., an employee ID transponder carried by an employee attempting to enter a suitable area) approaches and a personal identification process is initiated. In particular, when personal identification is performed to verify authorization, such a transponder can implement an additional layer of security together with the NFC device, because the personal identification process is initiated only when the person possessing such a transponder approaches. For example, if personal identification is initiated in this case and the personal identification is successful, a gate or door can be opened.

[0053] It should be noted that the list provided here is merely illustrative, and other recognition devices are similarly capable of performing the same tasks.

[0054] In another embodiment, the device for iris recognition during passage further includes a data memory having a database that stores the biometric features of the periocular facial region of persons with access rights. The control unit is configured to verify the personal identifier of the person to be identified through matching with the database.

[0055] The data memory may be, for example, part of the control unit or independent of it, but capable of communicating with the control unit. The data memory may also reside in the cloud, allowing access to a central database from various locations. The database may include, for example, data on the person's iris characteristics, or other biometric data, particularly data on the periorbital facial region. Preferably, the database is encrypted to prevent or at least make unauthorized access difficult. [Brief explanation of the drawing]

[0056] [Figure 1]This diagram shows a device for iris recognition during passage, positioned just before a user to be identified enters a detection corridor. In this case, the height settings of the first and second iris cameras are too high to detect the facial region around the user's eyes. [Figure 2] This figure shows the apparatus as it appears immediately after the person to be identified has entered the detection corridor and the height of the iris camera has been adjusted. [Figure 3] Figures 1 and 2 show the apparatus after the height of the iris camera has been adjusted to match the height of the periorbital facial region of the person to be identified, while the second iris camera is capturing an image of the periorbital facial region. [Figure 4] This figure shows exemplary images of the periorbital facial region captured by the first and second iris cameras. [Modes for carrying out the invention]

[0057] Detailed description of exemplary embodiments The depictions in each figure are approximate and not to scale. In the following descriptions of the figures, when the same reference numeral is used in different figures, that reference numeral represents the same or similar element; however, in some cases, the same or similar element is assigned a different reference numeral.

[0058] Figure 1 shows an exemplary device 100 for iris recognition during passage. The device 100 is used to identify a person 80 to be identified by detecting the periocular facial region 81 (see Figure 4) of the person 80 to be identified.

[0059] The device 100 includes a scene camera 30, a first iris camera 10, a second iris camera 20, a stroboscope lamp 60, a control unit 50 equipped with memory 51, and a recognition device 90 which can be configured as, for example, a photointerrupter 91 or an NFC device 92 (or other suitable device, such as a motion sensor or infrared distance sensor). The first iris camera 10, the second iris camera 20, and the stroboscope lamp 60 are further mounted on a common height-adjustable holder 40. Optionally, a scene camera may also be mounted on the height-adjustable holder 40. The first iris camera 10, the second iris camera 20, the scene camera 30, the stroboscope lamp 60, the recognition device 90, and the height-adjustable holder 40 are each electronically communicated with the control unit 50. The control unit 50 receives and / or transmits signals to each of these devices 10, 20, 30, 40, 60, and 90, thereby controlling the operation of the device 100.

[0060] The first iris camera 10 is a camera having a first field of view 11 and a fixed first focal plane 12, and the second iris camera 20 is a camera having a second field of view 21 and a fixed second focal plane 22. The first iris camera 10 has a first depth of field 13, and the second iris camera 20 has a second depth of field 23.

[0061] A "fixed" focal plane generally refers to a focal plane that cannot be shifted. Therefore, the first iris camera 10 and the second iris camera 20 are not particularly refocusable cameras, nor are they particularly plenoptic cameras (light field cameras) or cameras equipped with liquid lenses, but rather cameras with fixed focal planes.

[0062] The camera's field of view, as a whole, defines the angular range that can be detected by each camera. Therefore, the first field of view 11 defines the angular region detected by the first iris camera 10, and the second field of view 21 defines the angular region detected by the second iris camera 20.

[0063] The focal plane 12 of the first iris camera 10 and the focal plane 22 of the second iris camera 20 define the distance in front of each iris camera 10,20 at which the sharpest image is formed. Generally, the focal plane of a camera coincides with its respective focal length.

[0064] The depth of field 13 and 23 of the iris cameras 10 and 20, as explained above as general information regarding cameras, describe the size of the distance range in front of each camera in which an object is imaged clearly. Generally, the regions defined by the depth of field 13 and 23 extend from the plane between the corresponding focal planes 12 and 22 of the corresponding iris cameras 10 and 20 to a plane located further outward from the corresponding focal planes 12 and 22 of the corresponding iris cameras 10 and 20 (i.e., a plane further away from the camera). In other words, the depth of field defines the region around the focal planes 12 and 22 of the iris cameras 10 and 20 in which a sufficiently clear image is formed, enabling the extraction of image information necessary for iris recognition or recognition of other biological features of the periocular facial region 81.

[0065] Therefore, the first focal plane 12 and the first depth of field 13 (along with the first field of view 11, of course) define the first detection volume 14 of the first iris camera 10. Similarly, the second focal plane 22 and the second depth of field 23 (along with the second field of view 21) define the second detection volume 24 of the second iris camera 20. Thus, the first detection volume 14 and the second detection volume 24 define a spatial volume in front of the corresponding iris cameras 10 and 20, respectively, that enables the capture of a sufficiently clear image for iris recognition.

[0066] The first detection volume 14 and the second detection volume 24 are arranged in order from front to back in Figure 1 (and Figures 2 and 3). This allows for clear image acquisition along a relatively wide area along the detection corridor 70 using iris cameras 10 and 20 with fixed focal planes 12 and 22 (i.e., focal lengths). A complex refocusable camera system, such as a plenoptic camera, which provides a high data rate, is not used here. Because such a refocusable camera system is not used, the frame rate of the iris cameras 10 and 20 can be increased, thereby enabling accurate iris recognition over a wide area along the detection corridor 70 when passing through it, using simple iris cameras 10 and 20.

[0067] In Figures 1 to 3, the first detection volume 14 and the second detection volume 24 are arranged sequentially and spaced apart from each other. However, it should be noted that the two detection volumes 14 and 24 can be partially overlapped to enable continuous, clear image detection. Furthermore, it should be noted that the device with two iris cameras 10 and 20 is merely an example, and three or more iris cameras 10 and 20 can be provided to further expand the overall detection volume and thereby increase the success rate of personal identification.

[0068] The scene camera 30 has a third field of view 31 that is larger than the first field of view 11 of the first iris camera 10 and the second field of view 21 of the second iris camera 20. Therefore, the scene camera 30 can detect a larger angular range in front of the camera than the iris cameras 10 and 20, and is used to roughly orient the device 100 toward the periocular facial region 81, particularly during the personal identification process, as will be described later.

[0069] A relatively small field of view (i.e., a relatively small detection angle range in front of the camera) generally has small image distortion (as described above) and provides only a relatively small image section. Furthermore, a small field of view has the advantage that it is possible to use an image sensor with a relatively low pixel density, yet the required pixel resolution can still be obtained. In this case as well, the first iris camera 10 and the second iris camera 20 are used to capture the periocular facial region 81 for iris recognition and thus for personal identification, and here, the smallest possible image distortion and high resolution are desirable in order to extract the information necessary for personal identification from the image data. Therefore, the first iris camera 10 and the second iris camera 20 have smaller fields of view 11, 21 than the scene camera 30.

[0070] After the personal identification process is initiated, the scene camera 30 captures a scene image (preferably two-dimensional) containing the person 80 to be identified and transmits the corresponding image data to the control unit 50. The control unit 50 can then use a suitable image evaluation algorithm and known geometry of the camera system to determine / identify the height 82 of the periocular facial region 81 of the person 80. For this purpose, the control unit 50 can recognize, for example, typical structures in the image for the periocular facial region 81. For example, the memory 51 of the control unit 50 may contain corresponding comparison data, and the control unit 50 can also use machine learning algorithms. However, all other possible suitable image evaluation algorithms are also possible. Height determination can also be performed, for example, based on a calibration or mathematical model stored in the data memory 51 that allows the identified position on the scene image to be associated with a predetermined height.

[0071] Figure 1 shows the device 100 along with person 80 before the personal identification process for person 80 to be identified begins. Here, the heights of the first iris camera 10 and the second iris camera 20 (i.e., the heights of detection volumes 14 and 24 in particular) are too high, and the periocular facial region 81 of person 80 cannot be detected.

[0072] The personal identification process can be performed, for example, manually, by operating a corresponding switch or button, or automatically by the recognition device 90 as shown in Figures 1 to 3. If the recognition device 90 is, for example, a photointerrupter 91 or an NFC device 92, the personal identification process can be initiated when a person 80 passes the photointerrupter 91, or, for example, when a person 80 is carrying a transponder recognizable by an NFC device 92 and the person 80 carrying the transponder passes by the NFC device.

[0073] Since the iris cameras 10 and 20 have relatively small fields of view 11 or 21, in order to detect an image of the periorbital facial region 81, the detection height of the iris cameras 10 and 20 must be aligned with the height 82 of the periorbital facial region 81. For this purpose, after the person 80 enters the detection corridor 70, or even before the person 80 enters the detection corridor 70, the scene camera 30 first takes one or more images, and the control unit 50 determines the height 82 of the periorbital facial region 81 based on the image data as described above. Accordingly, the control unit 50 adjusts the heights of the first iris camera 10 and the second iris camera 20 to match the height 82 of the periorbital facial region 81, thereby enabling the iris cameras 10 and 20 to detect an image of the periorbital facial region 81. For example, the first iris camera 10 and the second iris camera 20 can be mounted on a height-adjustable holder 40 (not shown in detail), which may be, for example, a corresponding linear actuator or other device for shifting the iris cameras 10 and 20 in the height direction.

[0074] In Figures 1 to 3, an optional stroboscope lamp 60 is provided, which is configured to illuminate the periocular facial region 81 during image acquisition by the iris cameras 10 and 20 to stimulate the pupil response, thereby making iris features clearly recognizable and extractable in the image. In Figures 1 to 3, the stroboscope lamp 60 is also fixed to a height-adjustable support 40, and its height can be shifted together with the iris cameras 10 and 20. However, this is not essential. The stroboscope lamp 60 may, for example, be fixed in position to illuminate the entire detection corridor 70.

[0075] The scene camera 30 may be mounted at a fixed height, or it may be height-adjustable as described above, and may be height-adjustable together with the iris cameras 10 and 20 when their heights are adjusted. In such an arrangement, the control unit 50 takes into account the height of each scene camera 30 when determining the height. The scene camera 30 does not necessarily have to be mounted in direct proximity to the iris cameras 10 and 20 (as shown in Figures 1 to 3), and can be placed in other locations. The important thing is that the scene camera 30 can photograph the detection corridor 70 and that the field of view of the scene camera 30 enables the determination of the height of the periocular facial region 81.

[0076] Furthermore, it should be noted that the detection corridor 70 does not necessarily have to be a physically separated area as shown in Figures 1 to 3, and as a whole describes the detection area of ​​the device 100. A physical barrier is not necessarily required for this purpose. However, during personal identification, the person 80 must be facing the iris cameras 10 and 20 so that the iris cameras 10 and 20 can capture images of the periocular facial area 81 in at least the detection volume 14 and 24 area.

[0077] In Figures 2 and 3, some elements and reference numerals, such as the control unit 50 with memory 52 and the height-adjustable holder 40 in Figure 1, have been omitted for clarity.

[0078] Figure 2 shows the apparatus 100 of Figure 1, as described with reference to Figure 1, initiating the personal identification process after the person 80 to be identified enters the detection corridor 70, and adjusting the first iris camera 10 and the second iris camera 20 to the height 82 of the periocular facial region 81. In Figure 2, the person 80 to be identified is still located in front of the first detection volume 14 and the second detection volume 24, but is moving toward the detection volumes 14 and 24. Optionally, in this section, no further operation of the apparatus 100 is performed for the time being (after the height adjustment of the iris cameras 10 and 20 and, if applicable, the height adjustment of the stroboscope lamp 60 is completed).

[0079] Figure 3 shows the apparatus 100 after the person 80 to be identified has entered the second detection volume 24. From at least this point, the control unit 50 instructs the first iris camera 10 and the second iris camera 20 to begin image detection. However, the control unit 50 can already perform continuous image data detection by the iris cameras 10 and 20 from the start of the second section shown in Figure 2 (i.e., after the person 80 has entered the detection corridor 70). This eliminates the need for further localization of the person (e.g., by another photointerrupter or other means) to calculate when to start image acquisition. In this case, the control unit can discard images that do not have, for example, a recognizable periocular facial region 81. Furthermore, the control unit 50 instructs the stroboscope lamp 60 to illuminate the detection volumes 14 and 24 with stroboscope-like light pulses. Thus, while the person 80 moves through the first detection volume 14 and the second detection volume 24, the corresponding iris cameras 10 and 20 capture images of the periocular facial region 81 and transmit them to the control unit 50. Here, the stroboscope lamp can be optionally synchronized with image detection by the iris cameras 10 and 20, so that a light pulse is always output when exactly one image is captured by one of the iris cameras 10 and 20. This ensures that each image is sufficiently illuminated and that iris features can be extracted from the images of the periocular facial region 81.

[0080] Next, the control unit 50 extracts iris features from images captured by the iris cameras 10 and 20 and compares these iris features with corresponding comparison data in a database in order to identify the personal identifier of the person 80 to be identified. The database may be, for example, stored in the data memory 51. However, the comparison data may be stored in, for example, a central cloud memory, which makes it usable by the device 100 in various locations. Independent of the memory format, the comparison data is preferably stored in an encrypted state to prevent misuse by unauthorized reading of the comparison data.

[0081] Furthermore, the device 100 may be integrated into a more comprehensive system. If personal identification is performed, for example, for access monitoring, the device 100 (or its control unit 50) can communicate with the door mechanism and, if the detected personal identification corresponds to a person with access rights, automatically open, for example, a door, gate, or gatepost. Any other applications for which personal identification should be performed are also possible.

[0082] Figure 4 shows a typical image of the periorbital facial region 81. In such images, the eyes, particularly the irises, are clearly identifiable, and as described above, these biological features can be used for personal identification of the person 80.

[0083] It should be noted that, in general, personal identification may be based not only on iris features, but also on other biological features of the periocular facial region 81, such as the corresponding structures of the ocular region, or at least in conjunction with them. [Explanation of symbols]

[0084] 10. First iris camera 11. First Viewpoint 12. First Focal Plane 13. First depth of field 14. First detection volume 20. Second Iris Camera 21. Second Perspective 22 Second Focal Plane 23. Second depth of field 24 Second detection volume 30 Scene Camera 31. The Third Perspective 40 Height-adjustable holder 50 Control Unit 51 Data Memory 60 Stroboscope Lights 70 Detection Corridor 80 Persons to be identified 81 Periorocular Facial Region 82 Height of the periorbital facial region 90 Recognition device 91 Photo Interrupter 92 NFC devices 100 Devices for iris recognition

Claims

1. A device (100) for recognizing the iris when passing by, The aforementioned device (100) A first iris camera (10) having a first field of view (11) and a fixed first focal plane (12), A second iris camera (20) having a second field of view (21) and a fixed second focal plane (22), A scene camera (30) having a third field of view (31), Control unit (50) and Equipped with, The first iris camera (10) and the second iris camera (20) are height-adjustable. The first iris camera (10), the second iris camera (20), and the scene camera (30) are connected to the control unit (50). The control unit (50) is configured to use the scene camera (30) to detect an image of a person (80) to be identified that is located outside the combined detection area of ​​the first iris camera (10) and the second iris camera (20). The control unit (50) is configured to determine the height (82) of the facial region (81) around the eyes of the person (80) to be identified, based on the image detected by the scene camera (30). The control unit (50) is configured to adjust the heights of the first iris camera (10) and the second iris camera (20) in response to the height (82) of the identified periorbital facial region (81) so that the first iris camera (10) and the second iris camera (20) can detect the periorbital facial region (81). The control unit (50) is configured to detect the periorbital facial region (81) using the first iris camera (10) and the second iris camera (20), and to confirm the personal identification of the person (80) to be identified based on the detected periorbital facial region (81). A device (100) for recognizing the iris when passing by.

2. The device (100) for iris recognition during passage further includes a height-adjustable holding unit (40) that supports the first iris camera (10) and the second iris camera (20) and is connected to the control unit (50), The control unit (50) is configured to adjust the height of the first iris camera (10) and the second iris camera (20) by appropriately driving and controlling the height-adjustable holding unit (40). The device (100) for recognizing the iris when passing by, according to claim 1.

3. The device (100) for recognizing the iris when passing through further includes a stroboscope lamp (60) that operates in the near-infrared region. The control unit (50) is configured to drive and control the stroboscope lamp (60) during detection of the periorbital facial region (81) using the first iris camera (10) and the second iris camera (20) so that the stroboscope lamp (60) emits light pulses while detecting the periorbital facial region (81). A device (100) for recognizing the iris when passing through, according to claim 1 or 2.

4. The light pulse is synchronized with image detection by the first iris camera (10) and the second iris camera (20). The device (100) for recognizing the iris when passing by, according to claim 3.

5. The third field of view (31) is larger than the first field of view (11) and the second field of view (21). A device (100) for recognizing the iris when passing by, according to any one of claims 1 to 4.

6. The control unit (50) is configured to detect the periocular facial region (81) using the first iris camera (10) and the second iris camera (20) while the person to be identified (80) moves through the detection corridor (70). A device (100) for recognizing the iris when passing by, according to any one of claims 1 to 5.

7. The first iris camera (10) has a first depth of field (13), The second iris camera (20) has a second depth of field (23), A device (100) for iris recognition during passage, according to any one of claims 1 to 6.

8. The first focal plane (12) and the first depth of field (13) define the first detection volume (14), The second focal plane (22) and the second depth of field (23) define the second detection volume (24), The first detection volume (14) and the second detection volume (24) together form a combined detection area of ​​the first iris camera (10) and the second iris camera (20), each representing a spatial volume in front of the corresponding one of the first iris camera (10) and the second iris camera (20), and these spatial volumes are clearly imaged by the corresponding one of the first iris camera (10) and the second iris camera (20). The device (100) for recognizing the iris when passing by, according to claim 7.

9. The first detection volume (14) and the second detection volume (24) are at least partially different from each other. The device (100) for recognizing the iris when passing by, according to claim 8.

10. The first detection volume (14) and the second detection volume (15) are arranged in succession within the detection corridor (70) through which the person (80) to be identified passes during iris recognition. A device (100) for recognizing the iris when passing through, according to claim 8 or 9.

11. The first iris camera (10), the second iris camera (20), and the scene camera (30) are each non-plenooptic cameras. A device (100) for recognizing the iris when passing by, according to any one of claims 1 to 10.

12. The device (100) for recognizing the iris when passing through further comprises a recognition device (90), The recognition device (90) is connected to the control unit (50) and is configured to recognize the entry of the person to be identified (80) into the detection corridor (70) and to initiate the personal identification process as a response to said entry. A device (100) for recognizing the iris when passing by, according to any one of claims 1 to 11.

13. The recognition device (90) includes at least one of a photointerrupter (91), an NFC device (92), an infrared distance sensor, and a motion sensor attached to the entrance side of the detection corridor (70). The device (100) for recognizing the iris when passing by, according to claim 12.

14. The device (100) for iris recognition during passage further includes a data memory (51) having a database for storing the biological characteristics of the periocular facial region (81) of a person with access privileges. The control unit (50) is configured to detect the personal identifier of the person (80) to be identified through comparison with the database. A device (100) for recognizing the iris when passing by, according to any one of claims 1 to 13.

Images