Iris image acquisition system for expanding iris recognition range

Abstract

The disclosed content relates to an iris image acquisition system for expanding the iris recognition range, which enables the direction, zoom and focus of an iris capture camera for capturing iris images of distant objects to be quickly and easily achieved without the need for high-level information processing and calculation processing.

Description

Technical Field

[0001] The disclosed content relates to an iris image acquisition system for expanding the iris recognition range, which can quickly and accurately acquire the iris image of a distant object. Background Art

[0002] Unless otherwise stated in this specification, the contents described in this document do not constitute prior art to the claims of this application and should not be deemed to be prior art simply by being recorded in this section.

[0003] In recent years, there has been an increasing trend of using human information such as fingerprints and iris images in the security field.

[0004] In particular, the iris, as a part of the eye structure called the iris, can adjust the amount of light entering through the pupil like the aperture of a camera, and the iris is composed of the most complex and delicate fibrous tissue in the human body.

[0005] The iris forms a moiré-shaped iris pattern in early childhood and rarely changes throughout life. The iris pattern is different for each person, and even for twins and one's own left and right eyes. In addition, the iris is the only internal organ of the human body that can be recognized from the outside, and it is well known that it is protected inside the lens, so it is unlikely to be damaged by external influences.

[0006] As people realize that the uniqueness and persistence of iris patterns are most suitable for identifying human organs, iris recognition is increasingly used in the security field that relies on human body information identification.

[0007] Security technology using iris recognition is being tried not only at general entrances and exits that require close-range iris recognition, but also in security and access control areas such as entrances and exits with large traffic volumes (such as fast doors, airports, port facilities, etc.) that require long-range iris recognition.

[0008] As conventional systems generally applied to such long-distance iris recognition, systems developed by Sarnoff Corporation and Mitsubishi Corporation can be cited.

[0009] Among them, the portal system named "Iris On the Move" developed by Sarnoff Corporation uses two stationary short-focal-length megapixel-class high-resolution cameras to capture iris images of users approaching from a distance to pass through a door, and has the advantage of a faster processing speed (20 subjects / min).

[0010] However, with regard to the long-distance iris recognition system developed by Sarnoff Corporation, the system has limitations in terms of the position and posture of the user passing through the door. In particular, since the area captured by the camera is the entire area of ​​the door, a large number of infrared lighting devices are installed on the entire door in order to illuminate the entire shooting area, thereby increasing the system cost.

[0011] In addition, the long-distance iris recognition system developed by Mitsubishi Corporation is a system that detects the face of a user located far away within a wide range of motion by using a pan-tilt device, a wide-field-of-view (WFOV) camera, and a megapixel narrow-field-of-view (NFOV) camera, and acquires an iris image by using a distance prediction technology using the extracted facial feature information. The system has the advantage of having a wide range of motion due to the pan-tilt function.

[0012] However, in the case of the long-distance iris image acquisition system developed by Mitsubishi Corporation, since the system requires a large-scale near-infrared lighting device with high output characteristics that can fully illuminate the shooting area of ​​the narrow-angle camera at the megapixel level, it has the disadvantage of being difficult to control the large-scale lighting device used to achieve high output.

[0013] In addition, conventional long-distance iris image acquisition systems rely on well-known autofocus (AF) technology to achieve the focus of a zoom camera for iris photography. However, for conventional systems using zoom and autofocus (AF) technology, in order for the autofocus (AF) algorithm to find an image with a suitable focus, the subject needs to remain still for a certain period of time, and when the user remains still and waits for a few seconds, the performance of the system is stable, but when actually taking iris photos, there is a disadvantage that the activity of the subject needs to be strictly restricted.

[0014] In addition, although a high-performance autofocus (AF)-related computing processor is required to respond quickly to the movement of the subject, the following problem still exists: since the form of most acceleration chips is optimized for the visible light area, they are not suitable for iris photography that only performs high-resolution photography in the IR area.

[0015] For the above reasons, the previous iris recognition devices were implemented by close-range photography, and for systems that can perform long-range photography, they use expensive components and equipment and are implemented in a small-scale customization and supply manner. In addition, iris photography requires the use of technologies such as image analysis and high resource consumption in multiple stages, such as zoom and focus control of the lens and brightness control of the IR LED.

[0016] Recently, a long-distance iris image acquisition system has begun to be developed. This system introduces a face recognition camera to confirm the position of the iris, and controls the pan-tilt (PT) motor according to the face recognition position, so that the iris shooting camera is directed toward the eye.

[0017] As a solution to eliminate the above-mentioned problems, Korean Patent No. 10-0869998 (published on November 24, 2008) shows a long-distance iris image acquisition system, which includes: a wide-angle camera, which is used to capture the user's perceived image; a narrow-angle camera, which is a low-light camera or a near-infrared camera, which has a zoom lens and an autofocus function and captures the user's iris image; a near-infrared lighting device, which is arranged on the body of the narrow-angle camera and illuminates the shooting area of ​​the narrow-angle camera only when the narrow-angle camera is in operation; a PT (Pan-Tilt) device, which is used to make the optical axis approximately parallel to the horizontal axis of the camera. The wide-angle camera and the narrow-angle camera of the structural arrangement of the row are translated or swung; and a main processor, which translates or swings the PT device according to the eye position result detected after the face is recognized in the user perception image taken by the wide-angle camera, so that the eye position is located at the center of the image taken by the wide-angle camera, and then starts the near-infrared lighting device, and swings the narrow-angle camera until the eye is detected in the image taken by the narrow-angle camera, and after the eye is detected, the narrow-angle camera is translated or swung, when the user's eye is located at the center of the image taken by the narrow-angle camera, the zoom lens of the narrow-angle camera is adjusted to the zoom ratio required to obtain an iris image of a size suitable for iris recognition, and the iris image is obtained by automatic focusing.

[0018] In addition, Korean Patent Publication No. 10-2009-0130696 (published on December 24, 2009) shows a long-distance iris image acquisition system using translation and swing of a mirror, the system comprising: a wide-angle camera for capturing a user-perceived image; a narrow-angle camera having a function of adjusting a zoom lens and a focus lens for capturing a user's iris image; a near-infrared lighting device for illuminating a shooting area only when the narrow-angle camera is in operation; a mirror component for aligning the optical axes of the wide-angle camera and the narrow-angle camera by transmitting or reflecting visible light or transmitting or reflecting infrared light when the mirror component is arranged in front of the wide-angle camera and the narrow-angle camera; a mirror PT (Pan-Tilt) device for panning the wide-angle camera and the narrow-angle camera; and a mirror PT device for panning the wide-angle camera and the narrow-angle camera. A method for controlling the image quality of a wide-angle camera and a narrow-angle camera comprising: shifting or swinging the total reflection mirror to change the path of visible light or infrared light passing through the mirror component or reflected by the mirror component, thereby changing the shooting directions of the wide-angle camera and the narrow-angle camera with the same optical axes; and a main processor, which translates or swings the total reflection mirror of the mirror PT (Pan-Tilt) device according to the eye position result detected after recognizing the face in the user-perceived image shot by the wide-angle camera, so that the eye position is located at the center of the image shot by the wide-angle camera, and translates or swings the total reflection mirror of the mirror PT device while operating the near-infrared lighting device, and when the user's eye is located at the center of the image shot by the narrow-angle camera, adjusts the zoom lens of the narrow-angle camera to the zoom ratio required to obtain an iris image of a size suitable for iris recognition, and obtains the iris image by automatic focusing.

[0019] However, for the long-distance iris image acquisition system using a pan-tilt head (PT) as described above, since changing the zoom lens of the narrow-angle camera to the zoom ratio required to acquire an iris image of a size suitable for iris recognition requires high-level information processing and calculation processing capabilities, the system has the problem of taking a lot of time and money when actually acquiring a long-distance iris image. Summary of the invention

[0020] Technical issues

[0021] The present invention provides an iris image acquisition system for expanding the iris recognition range, which enables the direction, zoom and focus of an iris camera for capturing an iris image of a distant object to be quickly and easily realized without requiring high-level information processing and calculation processing based on the following geometric mapping information and distance information measured by a distance sensor. The above-mentioned geometric mapping information is generated based on the correlation between a wide-angle face recognition camera capable of recognizing the position of a face and eyes and a narrow-angle iris camera capable of controlling zoom-focus according to the distance.

[0022] In addition, the present invention is not limited to the technical tasks described above, and it is obvious that other technical tasks can be derived from the following description.

[0023] Workaround

[0024] An embodiment of an iris image acquisition system for expanding the range of iris recognition is shown herein. The iris image acquisition system for expanding the iris recognition range includes: a wide-angle face recognition camera, which can capture an object and detect the position of the face and eyes; a narrow-angle iris shooting camera, which has the function of adjusting a zoom lens and a focus lens, and is used to capture an iris image of the object; a near-infrared LED, which is used to emit near-infrared rays to a shooting area of ​​the iris shooting camera when the iris shooting camera is operated; a distance sensor, which is used to detect the distance to the object; a mapping information database, which is used to record and store geometric mapping information generated based on the correlation between the face recognition camera and the iris shooting camera according to the distance; a controller, which is used to control the operation of the face recognition camera, the iris shooting camera, the near-infrared LED and the distance sensor, and control the direction, zoom and focus of the iris shooting camera according to the following information: the eye position information of the object obtained through the image captured by the face recognition camera; the distance information to the object obtained through the sensor signal of the distance sensor; the geometric mapping information mapped to the above distance information in the mapping information database.

[0025] According to a preferred embodiment of the disclosed content, the iris shooting camera is set at a height of 1.3 m and is automatically swung within a range of 30° upward and 20° downward under the control of a controller, so that the iris image can be acquired regardless of the height of the subject.

[0026] According to a preferred embodiment of the disclosed content, the geometric mapping information in the mapping information database is obtained by parsing the homography information that changes according to the zoom magnifications of the face recognition camera and the iris shooting camera into a geometric homography matrix (Geometrical Homography) that reflects the spatial information.

[0027] According to a preferred embodiment of the disclosed content, the geometric mapping information in the mapping information database is generated according to the correlation coefficient of the face recognition camera and the iris shooting camera based on the distance to the object. The correlation coefficient is calculated in the following manner: generating first data for each distance to the object according to the zoom information and the focus information, extracting a plurality of reference points whose change rate of the focus information according to the distance exceeds a preset first critical value, dividing the first data into a plurality of intervals according to the plurality of reference points, and calculating the correlation coefficient according to the plurality of intervals.

[0028] According to a preferred embodiment of the disclosed content, the first data is generated by calculating zoom information for each distance to the object, calculating a depth range according to the zoom information and the distance to the object, and calculating focus information according to the depth range.

[0029] According to a preferred embodiment of the disclosed content, the plurality of intervals are arranged so that the depth range overlaps with a preset range.

[0030] Technical Effects

[0031] According to an iris image acquisition system for expanding the iris recognition range based on an embodiment of the disclosed content, since the system can be set up at entrances and exits with large traffic flows (such as fast doors, airports, port facilities, etc.), and acquire people's iris images at a long distance and perform authentication, it eliminates the inconvenience of previous commercial products that require the eyes to be close to the camera for personal authentication, and has the advantage of being able to quickly control the entry and exit of large numbers of people.

[0032] In addition, an iris image acquisition system for expanding the iris recognition range based on an embodiment of the disclosed content has the advantage that the direction, zoom and focus of the iris shooting camera can be quickly and easily realized based on the following geometric mapping information and the distance information measured by the distance sensor. The above-mentioned geometric mapping information is generated based on the correlation between the wide-angle face recognition camera and the iris shooting camera according to the distance to the object, wherein the wide-angle face recognition camera can estimate the position of the face and the eyes, and the iris shooting camera has the function of adjusting the zoom lens and the focus lens, so that the iris can be finely photographed.

[0033] In addition, an iris image acquisition system for expanding the iris recognition range based on an embodiment of the disclosed content has the advantage of being implementable even in resource-limited embedded systems by reducing the high-level information processing and calculation processes such as autofocus (AF) and image analysis performed when acquiring iris images of distant objects.

[0034] The effects of the present invention are not limited to the effects described above, and those skilled in the art can clearly understand other effects not mentioned herein based on the contents described in the claims. BRIEF DESCRIPTION OF THE DRAWINGS

[0035] Figure 1 is an overall block diagram of an iris image acquisition system for expanding the iris recognition range according to an embodiment of the disclosed content;

[0036] Figure 2It is a diagram of an embodiment of generating geometric mapping information based on the correlation relationship of the face recognition camera and the iris shooting camera according to the distance in the iris image acquisition system for expanding the iris recognition range according to one embodiment of the disclosed content;

[0037] Figure 3 is a detailed block diagram of an iris image acquisition system for expanding the iris recognition range according to an embodiment of the disclosed content;

[0038] Figure 4 is an example diagram of analyzing calibration data for correcting zoom lens tolerances according to one embodiment of the disclosed content;

[0039] Figure 5 is a diagram of an embodiment of analyzing zoom information and focus information according to a distance according to an embodiment of the disclosed content;

[0040] Figure 6 is a diagram of an embodiment in which intervals are arranged to overlap according to an embodiment of the disclosed content;

[0041] Figure 7 is a diagram illustrating a concept of calculating a correlation coefficient according to one embodiment of the disclosed content;

[0042] Figure 8 FIG. 1 is a diagram showing an embodiment of an iris image acquisition system for expanding an iris recognition range according to an embodiment of the disclosed content.

[0043] Description of Reference Numerals

[0044] 10: Face recognition camera

[0045] 20: Iris camera

[0046] 30:Near infrared LED

[0047] 40: Distance sensor

[0048] 50: Mapping Information Database

[0049] 60: Controller DETAILED DESCRIPTION

[0050] The following will describe the configuration, operation and effects of the preferred embodiments with reference to the accompanying drawings. For reference, in the following drawings, for the sake of convenience and clarity, the various components are omitted or generally illustrated, and the size of each component does not reflect the actual size. In addition, throughout the specification, the same reference numerals indicate the same components, and the reference numerals for the same components are omitted in individual drawings.

[0051] Figure 1It is an overall block diagram of an iris image acquisition system for expanding the iris recognition range according to an embodiment of the disclosed content.

[0052] According to an embodiment of the disclosed content, the iris image acquisition system for expanding the iris recognition range is not only set up at general entrances and exits where iris recognition is required, but also set up at entrances and exits with large traffic (such as fast doors, airports, port facilities, etc.), so as to be used as part of security equipment. Figure 1 As shown, the system includes: a wide-angle face recognition camera 10, which can capture an object and estimate the position of the face and eyes; a narrow-angle iris capture camera 30, which has the function of adjusting a zoom lens and a focus lens and is used to capture an iris image of the object; a near-infrared LED 30, which emits near-infrared rays to a capture area of ​​the iris capture camera 20 when the iris capture camera 20 is in operation; a distance sensor 40, which is used to detect the distance relative to the object; a mapping information database 50, which is used to record and store geometric mapping information generated based on the correlation relationship between the face recognition camera 10 and the iris capture camera 20 according to the distance; and a controller 60, which is used to control the operation of the face recognition camera 10, the iris capture camera 20, the near-infrared LED 30 and the distance sensor 40, and control the direction, zoom and focus of the iris capture camera 20 according to the following information: the object's eye position information obtained through the acquired image of the face recognition camera 10, the distance information relative to the object obtained through the sensor signal of the distance sensor 40, and the geometric mapping information mapped to the distance information in the mapping information database 50.

[0053] The face recognition camera 10 can estimate the positions of the face and eyes of a person by photographing the object, and is composed of a camera that photographs visible light images.

[0054] Since the face recognition camera 10 is used to capture a perceived image of an object and provide image data for estimating the position of a face and eyes, it is preferably a wide-angle camera having a lens with a viewing angle of 30° or more attached to a conventional color camera, and although it preferably has a general zoom function and a swing function, it may also be composed of a fixed camera that does not have a general zoom function and a swing function.

[0055] The face recognition camera 10 is preferably set at a height of 1.3 m so that it can acquire images containing faces regardless of the height of the subject, and can automatically swing within a range of 30° upward and 20° downward under the control of a controller 60 to be described later.

[0056] The iris camera 30 is disposed at a certain distance from one side, the upper part or the lower part of the face recognition camera 10. The iris camera 30 has the function of adjusting the zoom lens and the focus lens, so it plays the role of photographing the iris image of the subject.

[0057] Although the iris camera 30 may be a low-light camera (e.g., a VGA-class low-light camera) that obtains a brighter image than a general camera by amplifying a small amount of electronic signals generated in a CCD sensor when receiving a small amount of light, it is preferably a near-infrared camera that is more responsive to infrared wavelengths than general surveillance cameras. In addition, the iris camera 30 is preferably a narrow-angle camera with a zoom lens having a viewing angle of 1° to 5° attached thereto.

[0058] In addition, the iris camera 30 is preferably set at a height of 1.3 m so that the iris image can be acquired regardless of the height of the subject, and can be automatically swung within a range of 30° upward and 20° downward under the control of the controller 60 .

[0059] In addition, the face recognition camera 10 and the iris photographing camera 30 are arranged to form a parallel line of sight or to form an intersection outside the farthest distance to be photographed.

[0060] The iris camera 30 is provided with a near infrared LED 30. The near infrared LED 30 emits near infrared rays to the photographing area of ​​the iris camera 30 to illuminate the photographing area of ​​the iris camera 30 when the iris camera 30 is in operation.

[0061] The near infrared LED 30 is attached to the iris capturing camera 20 so as to work in conjunction with the iris capturing camera 30 , and lights up only when the iris capturing camera 30 is in operation.

[0062] The near-infrared LED 30 preferably has an illumination angle sufficient to illuminate the imaging area of ​​the iris imaging camera 30. For example, for an iris imaging camera 30 with a narrow angle of 5° at its maximum viewing angle, a near-infrared LED 30 with an illumination angle of 5° is suitable.

[0063] The near infrared LED 30 can be attached in a single number to any of the upper, lower, left, and right positions of the iris camera 30 body, and light up only when the iris camera 30 is in operation. In this case, it is preferable to make the reflected light caused by the illumination converge above or below the iris which is not used for iris recognition.

[0064] In addition, two near-infrared LEDs 30 may be installed one above and one below or on the left and right sides of the main body of the iris camera 30, and light up only when the iris camera 20 is in operation. In this case, the irradiation direction of the two near-infrared LEDs 30 should be inclined toward the shooting area based on the center line, and since the overlapping area increases with the increase of distance, it is preferably designed to fully include the area to be shot by the iris camera 30.

[0065] Generally speaking, in order to capture near-infrared images at a long distance, high-output energy is required, which is related to the output performance of the system, and therefore, the cost increases as the distance increases. To solve the above problem, the near-infrared LED 30 is preferably controlled by the controller 60 to irradiate the area corresponding to the iris position with near-infrared rays only during the time of a few milliseconds (ms) or less during the capture.

[0066] The distance sensor 40 is disposed on the body of the face recognition camera 10 or the iris capture camera 30. The distance sensor 40 measures the distance between the face recognition camera 10 or the iris capture camera 30 and the object, and transmits a sensor signal for calculating distance information to the controller 60. The distance sensor may be composed of a conventional ultrasonic distance sensor.

[0067] In addition, the controller 60 for controlling the operations of the above-mentioned face recognition camera 10 , iris photographing camera 30 , near infrared LED 30 , and distance sensor 40 includes a mapping information database 50 .

[0068] The mapping information database 50 is formed to record and store geometric mapping information generated based on the correlation relationship according to the distance between the face recognition camera 10 and the iris imaging camera 20 . Figure 2 An embodiment of generating geometric mapping information according to the correlation relationship of distance based on a face recognition camera and an iris capture camera is shown.

[0069] like Figure 2 As shown, the geometrical mapping information in the mapping information database 50 can be obtained by parsing the homography information of the face recognition camera 10 and the iris capture camera 30 that changes according to the zoom factor into a geometrical homography matrix (Geometrical Homography) that reflects the spatial information.

[0070] Specific methods for solving the actual distance and size of the object to be found in the actual image relative to the camera include: a method of directly solving it based on geometry using hands and a pen, a method using a homography matrix, and a method using 3D transformation.

[0071] Among them, homography refers to a certain transformation relationship between corresponding points when projecting a plane onto another plane. Homography is usually represented by a 3×3 matrix and is defined as a transformation relationship that is established for the homogeneous coordinate expression of corresponding points. Homography is not only established between two planes with a direct projection relationship, but also generally between all planes that are directly or indirectly connected due to the projection relationship.

[0072] The method of using the homography matrix to obtain the actual position of the object detected in the image is as follows:

[0073] First, determine any four points on the ground within the camera's field of view, measure the ground coordinates of these four points through actual measurement, and set the reference (origin, etc.) of the ground coordinate system. After measuring the ground coordinates of the four points according to the set reference (ground coordinate system), find the image coordinates of each point (pixel coordinates of the camera image). You can use the findHomogray function of opencv or use the following code to directly calculate the homography transformation matrix H from pixel coordinates to ground coordinates. Once the homography matrix H is obtained, the subsequent process is simple. The ground coordinates corresponding to any image pixel coordinate p(x,y) are obtained by multiplying H by (x,y,1) which expands p to homogeneous coordinates and re-transforming the result into 2d coordinates. That is, if H*(x,y,1) T =(a,b,c) T , then the ground coordinates to be solved are (a / c,b / c).

[0074] Such geometric homography interpretation methods are well known in the art, and therefore, for simplicity, further detailed description is omitted here.

[0075] The face recognition camera 10, iris camera 30, near infrared LED 30 and distance sensor 40 are connected to the controller 60 through wired or wireless communication. The controller 60 controls the operation of the face recognition camera 10, iris camera 30, near infrared LED 30 and distance sensor 40, and includes the mapping information database.

[0076] In particular, the controller 60 controls the direction, zoom and focus of the iris capture camera 30 based on the following information: eye position information, which is obtained by acquiring an image through the face recognition camera 10; distance information from the object, which is obtained through the sensor signal of the distance sensor 40; and geometric mapping information, which is geometric mapping information of the mapping information database 50 mapped to the above distance information.

[0077] The iris image acquisition system for expanding the iris recognition range according to an embodiment of the disclosed content may be composed of a main body frame in which the face recognition camera 10 and the iris photographing camera 30 are set to be swingable, and the controller 60 including the mapping information database 50 may also be installed on the main body frame. According to the embodiment, the face recognition camera 10 and the iris photographing camera 30 are respectively arranged to be swingable and spaced apart, and may also be connected to the controller 60 including the mapping information database 50 based on wired or wireless communication.

[0078] Figure 3 is a detailed block diagram of an iris image acquisition system for expanding the range of iris recognition according to an embodiment of the disclosed content.

[0079] refer to Figure 3 Observe the operation flow of the iris image acquisition system for expanding the iris recognition range according to one embodiment of the present invention. First, the distance information from the object for iris recognition can be extracted through the distance sensor 40, and the controller 60 can control the operation of the near-infrared LED 30 and the iris camera 20 based on the geometric mapping information and the distance information of the mapping information database 50, and can especially control the zoom and focus of the iris camera 20. At this time, as described above, one embodiment of the present invention does not rely on the autofocus function, but can perform instant focusing based on the mapping information, does not require high-level calculations, and can perform fast processing.

[0080] In addition, the iris image acquisition system for expanding the iris recognition range according to an embodiment of the present invention shoots the subject through the wide-angle face recognition camera 10, extracts the eye position based on the face feature algorithm, and enables the iris shooting camera 20 to accurately shoot the eye position of the subject through coordinate mapping.

[0081] Figure 4 is an example diagram of analyzing calibration data for correcting zoom lens tolerances according to one embodiment of the disclosed content, Figure 5 is a diagram of an embodiment of analyzing zoom information and focus information according to distance according to an embodiment of the disclosed content, and Figure 6 is a diagram of an embodiment in which intervals are arranged to overlap according to an embodiment of the disclosed content.

[0082] The geometric mapping information of the mapping information database 50 according to one embodiment of the disclosed content may be generated based on a correlation coefficient according to a distance from an object based on the face recognition camera 10 and the iris photographing camera.

[0083] refer to Figures 4 to 6At this time, the correlation coefficient can be calculated in the following manner: first data is generated for various distances from the object according to the zoom information and the focus information, multiple reference points are extracted according to the focus information based on the distance change rate exceeding the preset first critical value, the first data is divided into multiple intervals according to the multiple reference points, and the correlation coefficient is calculated for each of the multiple intervals.

[0084] The reason for generating the first data for various distances is that the shortest distance for iris recognition differs depending on the distance to the object, and therefore, it is preferable to generate the first data for various distances to the object based on the zoom information and the focus information.

[0085] At this time, in order to quickly process the zoom information and focus information based on the distance, the first data can be divided into multiple intervals, and the correlation coefficient can be calculated for each interval. At this time, the reason for setting the reference point for dividing the intervals based on the rate of change of the focus information according to the distance is to ensure the minimum clarity for identifying the iris.

[0086] In addition, the first data may be generated by calculating zoom information according to the distance to the object, calculating a depth range according to the zoom information and the distance to the object, and calculating focus information according to the depth range. The focus information is calculated so that the object is included in the camera depth.

[0087] In addition, as described above, a plurality of intervals are set so that the depth range overlaps with a preset range, and thus there is no position that is missed depending on the distance to the object.

[0088] Figure 7 is a diagram showing a concept of calculating a correlation coefficient according to one embodiment of the disclosed content, and Figure 8 FIG. 1 is a diagram showing an embodiment of an iris image acquisition system for expanding an iris recognition range according to an embodiment of the disclosed content.

[0089] refer to Figure 7 One embodiment of the disclosed content can generate mapping information by accumulating distance-based correlation coefficients of the face recognition camera 10 and the iris capture camera 20. Thus, iris recognition can be performed faster.

[0090] Figure 8 is an example of estimating the iris position of the subject using the correlation coefficient. According to one embodiment of the disclosed content, eye position information is first extracted from the face image of the subject captured by the face recognition camera 10, and the iris capturing camera 20 is controlled based on the eye position information to capture the eyes of the subject, and to confirm the position of the iris, the zoom and focus of the iris capturing camera 20 are controlled based on the distance information from the subject and the mapping information, thereby extracting the eye image of the subject.

[0091] As described above, for an iris image acquisition system for expanding the iris recognition range according to an embodiment of the disclosed content, the system can be set up at entrances and exits with large traffic flows (such as fast doors, airports, port facilities, etc.), and acquire a person's iris image at a long distance and perform authentication, thereby eliminating the inconvenience of previous commercial products that require the eyes to be close to the camera for personal authentication, and can quickly control the entry and exit of a large number of people.

[0092] In addition, for an iris image acquisition system for expanding the iris recognition range according to an embodiment of the disclosed content, the system can quickly and easily realize the direction, zoom and focus of the iris shooting camera according to the following geometric mapping information and the distance information measured by the distance sensor 40, wherein the above-mentioned geometric mapping information is generated based on the correlation relationship according to the distance between the wide-angle face recognition camera 10 capable of estimating the position of the face and eyes and the iris shooting camera 20 having the function of adjusting the zoom lens and the focus lens so as to be able to take a fine picture of the iris.

[0093] In addition, according to an iris image acquisition system for expanding the iris recognition range based on an embodiment of the disclosed content, since the system reduces the higher-level information processing and calculation processes such as autofocus (AF) and image analysis performed when acquiring iris images of distant objects, it can be implemented even in embedded systems with limited resources.

[0094] Although the preferred embodiments of the present invention are described above with reference to the accompanying drawings, the embodiments described in the specification and the constituent elements illustrated in the accompanying drawings are limited to the most preferred embodiment of the present invention and do not represent the entire technical concept of the present invention. Therefore, it should be understood that at the time of submitting this application, there may be equivalents and modified examples that can replace the embodiments. Therefore, it should be understood that the embodiments described above are not exemplary and restrictive in all aspects. The scope of the present invention is shown by the claims rather than by the contents of the specification, and the definition and scope of the claims and all changes or deformations derived from their equivalent concepts should be interpreted as being included in the scope of the present invention.

[0095] Industrial application possibilities

[0096] The disclosed content may be applicable to the field of camera devices or security systems based on acquiring iris images of a subject.

Claims

1. An iris image acquisition system for expanding the iris recognition range, comprising: A wide-angle facial recognition camera that estimates the position of faces and eyes by photographing the subject; A narrow-angle iris-capturing camera having a function of adjusting a zoom lens and a focus lens and used to capture an iris image of a subject; a near-infrared LED that emits near-infrared rays toward a photographing area of ​​the iris photographing camera when the iris photographing camera is in operation; a distance sensor, which measures the distance to an object; A mapping information database, which records and stores geometric mapping information generated based on the correlation between the face recognition camera and the iris camera according to the distance; as well as a controller that controls the operations of the face recognition camera, the iris camera, the near infrared LED, and the distance sensor, and controls the direction, zoom, and focus of the iris camera based on the eye position information of the object obtained through the acquired image of the face recognition camera, the distance information to the object obtained through the sensor signal of the distance sensor, and the geometric mapping information mapped to the distance information in the mapping information database, The geometric mapping information in the mapping information database is generated based on a correlation coefficient based on the distance between the face recognition camera and the iris photography camera and the object. The correlation coefficient is calculated by: generating first data for each distance from the object according to zoom information and focus information; extracting a plurality of reference points whose change rate of the focus information according to the distance exceeds a preset first critical value; dividing the first data into a plurality of intervals according to the plurality of reference points; and calculating the correlation coefficient according to the plurality of intervals, The first data is generated by: calculating the zoom information for each distance from the object; calculating a depth range according to the zoom information and the distance from the object; and calculating the focus information according to the depth range, The plurality of intervals are set such that the depth range overlaps with a preset range.

2. The iris image acquisition system for expanding the iris recognition range according to claim 1, wherein: The iris photographing camera is set at a height of 1.3 m, and can be automatically swung within a range of 30° upward and 20° downward under the control of the controller, so that an iris image can be acquired regardless of the height of the subject.

Images