Biometric determination device and biometric determination method

Abstract

To provide a biometric determination device and a biometric determination method capable of improving the accuracy of determining fake images of a living body.SOLUTION: An information processing terminal comprises a controller, a visible light camera, an infrared camera, and an infrared light source. The controller determines that a subject is a living body when determining that the subject's eye has pupil absorption and the subject's eye has corneal reflection on the basis of an infrared image of the subject's eye captured by the infrared camera, and determining that the subject's eye has pupil absorption and the subject's eye does not have corneal reflection on the basis of a visible light image of the subject's eye captured by the visible light camera.SELECTED DRAWING: Figure 1

Description

Technical Field

[0001] The present invention relates to a biological determination device and a biological determination method.

Background Art

[0002] In a face authentication device, there is a possibility that face authentication may be illegally permitted by using a fake image of a person or the like. In order to prevent impersonation using a fake image of a person or the like, the face authentication device performs a biological determination to determine whether the subject captured by the camera is a living body. As a method of biological determination, there is a method of determining a living body by using the biological characteristics of a person's eyeball. For example, the method described in Patent Document 1 uses an image captured by illuminating coaxially with the optical axis of a camera to detect a retinal reflection and determine it as a living body.

Prior Art Documents

Patent Documents

[0003]

Patent Document 1

Summary of the Invention

Problems to be Solved by the Invention

[0004] As advanced deepfake technology and generative models evolve, there is a possibility that in biological determination, highly sophisticated fake images of living bodies (humans) that reflect (imitate) the characteristics of images caused by the biological characteristics of the eyeballs may be generated and used. It is required that the biological determination device improve the determination accuracy for fake images of living bodies so that such highly sophisticated fake images of living bodies can be correctly determined as non-living bodies. The method described in Patent Document 1 may not be able to correctly determine a highly sophisticated fake image of a living body (for example, a highly sophisticated fake image of a living body that reflects a state where a retinal reflection occurs) as a non-living body.

[0005] The present invention has been made to solve the above problems. That is, one of the objects of the present invention is to provide a living body determination device and a living body determination method capable of improving the determination accuracy for a fake image of a living body.

Means for Solving the Problems

[0006] In order to solve the above problems, a living body determination device of the present invention includes an infrared light source that generates infrared light and illuminates the eyes of a subject that is a living body determination target, an infrared camera that images the subject, a visible light camera that images the subject, and a controller that controls the infrared camera and the visible light camera. The controller detects the eyes of the subject from the infrared image of the subject imaged by the infrared camera, performs a first pupil absorption determination to determine whether there is pupil absorption in the eyes of the subject, and a first corneal reflection determination to determine whether there is corneal reflection in the eyes of the subject. The controller also detects the eyes of the subject from the visible light image of the subject imaged by the visible light camera, performs a second pupil absorption determination to determine whether there is pupil absorption in the eyes of the subject, and a second corneal reflection determination to determine whether there is corneal reflection in the eyes of the subject. Based on the determination results of the first pupil absorption determination and the first corneal reflection determination, and the second pupil absorption determination and the second corneal reflection determination, the controller determines whether a living body determination condition, which is a condition indicating that the subject is a living body, is satisfied. When the living body determination condition is satisfied, the controller determines that the subject is a living body. When the living body determination condition is not satisfied, the controller determines that the subject is not a living body.

[0007] The biological determination device of the present invention includes an infrared light source that generates infrared light and illuminates the eyes of a subject that is the object of biological determination, a visible light source that generates visible light and illuminates the eyes of the subject, an infrared camera that images the subject, a visible light camera that images the subject, and a controller that controls the infrared camera and the visible light camera. The controller detects the eyes of the subject from the infrared image of the subject captured by the infrared camera, performs a first pupil absorption determination to determine whether there is pupil absorption in the eyes of the subject, and a first corneal reflection determination to determine whether there is a corneal reflection in the eyes of the subject. The controller also detects the eyes of the subject from the visible light image of the subject captured by the visible light camera, performs a second pupil absorption determination to determine whether there is pupil absorption in the eyes of the subject, changes the number of times the visible light source is turned on, and performs a pupil size determination to determine whether the size of the pupil changes when the number of times the visible light source is turned on is changed. Based on the determination results of the first pupil absorption determination and the first corneal reflection determination, and the second pupil absorption determination and the pupil size determination, the controller determines whether a biological determination condition, which is a condition indicating that the subject is a living body, is satisfied. When the biological determination condition is satisfied, the controller determines that the subject is a living body. When the biological determination condition is not satisfied, the controller determines that the subject is not a living body. It is configured as follows.

[0008] The biological determination method of the present invention is a biological determination method using an infrared light source that generates infrared light and illuminates the eyes of a subject that is the object of biological determination, an infrared camera that images the subject, a visible light camera that images the subject, and a controller that controls the infrared camera and the visible light camera. The controller detects the eyes of the subject from the infrared image of the subject captured by the infrared camera, and performs a first pupil absorption determination to determine whether there is pupil absorption in the eyes of the subject, and a first corneal reflection determination to determine whether there is corneal reflection in the eyes of the subject. The eyes of the subject are detected from the visible light image of the subject captured by the visible light camera, and a second pupil absorption determination to determine whether there is pupil absorption in the eyes of the subject, and a second corneal reflection determination to determine whether there is corneal reflection in the eyes of the subject are performed. Based on the determination results of the first pupil absorption determination and the first corneal reflection determination, and the second pupil absorption determination and the second corneal reflection determination, it is determined whether a biological determination condition, which is a condition indicating that the subject is a living body, is satisfied. When the biological determination condition is satisfied, it is determined that the subject is a living body, and when the biological determination condition is not satisfied, it is determined that the subject is not a living body.

[0009] The biological determination method of the present invention uses an infrared light source that generates infrared light and illuminates the eyes of a subject that is the object of biological determination, a visible light source that generates visible light and illuminates the eyes of a subject that is the object of biological determination, an infrared camera that images the subject, a visible light camera that images the subject, and a controller that controls the infrared camera and the visible light camera. The controller detects the eyes of the subject from the infrared image of the subject captured by the infrared camera, performs a first pupil absorption determination to determine whether there is pupil absorption in the eyes of the subject, and a first corneal reflection determination to determine whether there is corneal reflection in the eyes of the subject. The controller detects the eyes of the subject from the visible light image of the subject captured by the visible light camera, performs a second pupil absorption determination to determine whether there is pupil absorption in the eyes of the subject, changes the number of times the visible light source is turned on, and performs a pupil size determination to determine whether the size of the pupil changes when the number of times the visible light source is turned on. Based on the determination results of the first pupil absorption determination and the first corneal reflection determination, and the second pupil absorption determination and the pupil size determination, the controller determines whether a biological determination condition, which is a condition indicating that the subject is a living body, is satisfied. When the biological determination condition is satisfied, the controller determines that the subject is a living body. When the biological determination condition is not satisfied, the controller determines that the subject is not a living body.

Advantages of the Invention

[0010] According to the present invention, the determination accuracy for fake images of living bodies can be improved. Note that the effects described here are not necessarily limited, and any of the effects described in the present disclosure may be applicable.

Brief Description of the Drawings

[0011]

Figure 1

Figure 2

Figure 3A

Figure 3B

Figure 3C

Figure 4A

Figure 4B

Figure 4C

Figure 4D

Figure 5A

Figure 5B

Figure 5C

Figure 6

Figure 7

Figure 8

Figure 9A

Figure 9B

Figure 9C

Figure 10A

Figure 10B

Figure 10C

Figure 11A

Figure 11B

Figure 11C

Figure 12

Figure 13

Figure 14

Figure 15A

Figure 15B

Figure 15C

Figure 16A

Figure 16B

Figure 16C

Figure 17A

Figure 17B

Figure 17C

Figure 18

Figure 19

Figure 20

Figure 21

Mode for Carrying Out the Invention

[0012] Hereinafter, each embodiment of the present invention will be described with reference to the drawings. In all the drawings of the embodiments, the same or corresponding parts may be denoted by the same reference numerals. In the following description, when describing the processing with a functional block as the subject, the subject of the processing may be the CPU or the device instead of the functional block.

[0013] The biological determination device (information processing terminal 1) according to each embodiment described below is assumed to be installed, for example, in a facility (such as a store or an office) where consent for imaging has been obtained for the purpose of determining whether a subject is a living body.

[0014] Further, such a biological determination device (information processing terminal 1) is connected to a face authentication device, and by adding a biological determination function to such a face authentication device, it is possible to prevent impersonation of a person. However, technically, it is not limited to the above, and it can be connected to any device other than the face authentication device, particularly a device that needs to determine whether it is a living body (for example, a drowsiness monitoring system). Alternatively, it may be used alone as a biological determination device.

[0015] <<First Embodiment>> <Configuration> An information processing terminal 1 according to a first embodiment of the present invention will be described. FIG. 1 is a configuration diagram showing a hardware configuration example of the information processing terminal 1 according to the first embodiment of the present invention. FIG. 2 is a diagram showing a functional block of the controller 11 of the information processing terminal 1 in FIG. 1. As shown in FIG. 1, the information processing terminal 1 includes an infrared light source 10, a controller 11, a visible light camera 12, an infrared camera 13, and a display 14. Note that the information processing terminal 1 may be referred to as a "biometric determination device".

[0016] The infrared light source 10 is a light source that generates infrared rays and illuminates the subject 2, and is, for example, an infrared LED (light-emitting diode) or the like.

[0017] The controller 11 includes a CPU (Central Processing Unit) 111, a ROM (Read Only Memory) 112, a RAM (Random Access Memory) 113, a camera processing unit 114, a light source control unit 115, and a display control unit 116. These are connected to each other via a bus so as to be able to communicate information.

[0018] The CPU 111 loads the program stored in the ROM 112 into the RAM 113. The CPU 111 realizes various functions of the information processing terminal 1 by executing the program loaded into the RAM 113. The ROM 112 is a non-volatile storage medium. Various programs for realizing various functions of the information processing terminal 1 are stored in the ROM 112. The RAM 113 is a volatile storage medium. As described above, the program executed by the CPU 111 is loaded into the RAM 113, and data used when the CPU 111 executes the program is temporarily stored.

[0019] The camera processing unit 114 includes an input / output interface (not shown) and the like, and inputs (acquires) image data including two-dimensional information from the visible light camera 12 and the infrared camera 13 for each frame, and supplies such data to the CPU 111 and the like.

[0020] The light source control unit 115 includes an input / output interface (not shown) and controls the infrared light source 10 which is a light source. At this time, when there are a plurality of light sources, the light source control unit 115 may control each light source independently. Between the light source control unit 115 and the light source, data can be transmitted and received according to protocols such as USB (Universal Serial Bus), I2C (Inter-Integrated Circuit), SPI (Serial Peripheral Interface), and UART (Universal Asynchronous Receiver Transmitter).

[0021] The display control unit 116 includes an input / output interface (not shown) and controls the generation and display of an image (video). The display control unit 116 transmits a video signal indicating an image to the display 14 and displays the image on the display 14.

[0022] The visible light camera 12 is an imaging device that detects light with wavelengths in the visible light region and converts it into an electrical signal (video signal) to generate an image.

[0023] The infrared camera 13 is an imaging device that detects infrared rays and converts them into an electrical signal (video signal) to generate an image. From the perspective of realizing a simpler configuration in the information processing terminal 1, it is preferable that the optical axis of the infrared camera 13 and the optical axis of the infrared light source 10 are different from each other. As in Patent Document 1 (Japanese Unexamined Patent Application Publication No. 2005-040591), in order to make the optical axes of the camera and the light source the same, a mechanism for arranging a plurality of mirrors and lenses for folding light is required, which may increase the thickness and weight. Also, from the perspective of realizing a simpler configuration in the information processing terminal 1, it is preferable that the optical axis of the visible light camera 12 and the optical axis of the infrared light source 10 are different from each other.

[0024] The display 14 is a display device capable of displaying an image.

[0025] As shown in FIG. 2, the controller 11 includes an eye detection unit 102, a pupil absorption determination unit 103, a corneal reflection determination unit 104, a living body determination unit 105, a light source control unit 115, and a display control unit 116.

[0026] The eye detection unit 102 detects the eyes of the subject 2 from the images input from each of the visible light camera 12 and the infrared camera 13. For eye detection, for example, deep learning can be used. When using deep learning, it can be realized by previously training the eyes of a plurality of people with a deep neural network. Also, pattern matching can be used. As another method, when the positions and directions of the eyes of the subject 2 and the visible light camera 12 and the infrared camera 13 are fixed, without performing the above-described eye detection process, specific positions in the images captured by the visible light camera 12 and the infrared camera 13 can be detected as eyes.

[0027] The eye detection unit 102 outputs an image obtained by cutting out the detected eye region from the images input from the visible light camera 12 and the infrared camera 13 to the pupil absorption determination unit 103 and the corneal reflection determination unit 104. At this time, it is also possible to output an image obtained by detecting both the left and right eyes and cutting out the regions of both eyes, or to output an image obtained by detecting only one of the left and right eyes and cutting out the region of one eye.

[0028] The eye detection unit 102 cuts out so as to include at least the iris and the pupil. The cutting may be an outer circumscribed rectangle or an inner circumscribed rectangle of the iris region, or a circle adapted to the iris region. When cutting out an outer circumscribed rectangle or an inner circumscribed rectangle, for example, a certain ratio of the distance between the eyes centered on the detected eye position can be used as the width and height of the region. When cutting out a circle, a certain ratio of the distance between the eyes centered on the detected eye position can be used as the radius. Also, as another method, an edge filter such as a Laplacian filter can be used to obtain the outer shape of the iris region, and an inner circumscribed rectangle, an outer circumscribed rectangle, or the inside of the outer shape can be cut out.

[0029] The pupil absorption determination unit 103 determines whether light is absorbed by the pupil and outputs the determination result. The pupil absorption determination unit 103 takes the eye image output by the eye detection unit 102 as input and determines whether light is absorbed by the eye image.

[0030] As a method for determining whether light is absorbed by the pupil, it can be determined by whether there are dark pixels (dark pixels indicating that light is absorbed by the pupil) in the image.

[0031] The pupil absorption determination unit 103 scans the pixels in the eye region and can determine that light is absorbed by the pupil when there are a certain number or more of pixels below a certain value.

[0032] As another method for determining whether light is absorbed by the pupil, it can be determined based on the histogram of the pixel values of the pixels in a specific region of the eye image (for example, the inscribed rectangle of the iris region, etc.).

[0033] For example, as another method for determining whether light is absorbed by the pupil, there is also a method of determining from the shape of the histogram. Specifically, for example, first calculate the moving average of the histogram, differentiate the obtained moving average to obtain the maximum value. It can be determined that light is absorbed by the pupil from the number of obtained maximum values and the positions of the maximum values. For example, it can be determined under conditions such as having one or more maximum values below the pixel value of 50 and having one or more maximum values between the pixel values of 100 and 200.

[0034] As other methods, it is also possible to obtain the gradient, kurtosis, skewness, half-value width, etc. of the graph and use a prediction model such as a decision tree to determine that light is absorbed by the pupil. At this time, the gradient, etc. may be calculated in multiple ways by dividing the horizontal axis of the graph. As yet another method, deep learning can be used. When using deep learning, it can be realized by previously training the inscribed rectangle of the iris region of a real human in a deep neural network.

[0035] When the eyes detected by the eye detection unit 102 are both eyes, the pupil absorption determination unit 103 independently determines whether light is absorbed by the pupils respectively, and outputs the determination results independently.

[0036] The corneal reflection determination unit 104 determines whether the cornea reflects light and outputs the determination result. Since the cornea reflects light when the eyeball is irradiated with light of a certain intensity, the corneal reflection determination unit 104 determines the presence or absence (and the number of corneal reflections (in other words, "corneal reflection images")) of the corneal reflection, and uses this for biological determination. Note that this corneal reflection image is known as the Purkinje image.

[0037] The corneal reflection determination unit 104 takes the eye image output by the eye detection unit 102 as input, and determines whether light is reflected in the eye image.

[0038] As a method for determining whether the cornea reflects light, it can be determined by whether there are bright pixels (bright pixels indicating that the cornea reflects light) in the image.

[0039] The corneal reflection determination unit 104 scans the pixels in the eye region, and can determine that the cornea reflects light when there are a certain number or more of pixels above a certain value. Also, the corneal reflection determination unit 104 can obtain the number of corneal reflections by taking the number of parts where there are a certain number or more of pixels above a certain value as the number of corneal reflections.

[0040] As another method for determining whether the cornea reflects light, it can be determined based on the histogram of the pixel values of the pixels in a specific region of the eye image (for example, the inscribed rectangle of the iris region, etc.).

[0041] For example, as another method for determining whether the cornea reflects light, there is also a method of determining from the shape of the histogram. Specifically, for example, first calculate the moving average of the histogram, differentiate the obtained moving average to obtain the maximum value. It is possible to determine that the cornea reflects light from the number and position of the obtained maximum values. For example, it can be determined under conditions such as having one or more maximum values between pixel values of 100 to 200 and having one or more maximum values at pixel values of 230 or more. Further, the corneal reflection determination unit 104 can obtain the number of corneal reflections by using the number of maximum values as the number of corneal reflections.

[0042] As other methods, it is possible to obtain the gradient, kurtosis, skewness, full width at half maximum, etc. of the graph, and use a prediction model such as a decision tree to determine whether the cornea reflects light or obtain the number of corneal reflections. At this time, a plurality of gradients, etc. may be calculated by dividing the horizontal axis of the graph. As yet another method, deep learning can be used. When using deep learning, it can be realized by previously training an inscribed rectangle of the iris region of a real human in a deep neural network.

[0043] When the eyes detected by the eye detection unit 102 are both eyes, the corneal reflection determination unit 104 independently determines whether each cornea reflects light and independently outputs the determination result.

[0044] The living body determination unit 105 determines whether the subject 2 is a living body based on the determination results output by the pupil absorption determination unit 103 and the corneal reflex determination unit 104, and outputs the determination result. When the pupil absorption determination unit 103 and the corneal reflex determination unit 104 output the determination results of both eyes independently for each eye, the living body determination unit 105 may determine whether the subject 2 is a living body as follows. First, the living body determination unit 105 performs living body determination for each eye based on the determination results for the same eye output by the pupil absorption determination unit 103 and the corneal reflex determination unit 104, and then outputs a living body determination result according to the living body determination results of the left and right eyes respectively. At this time, it may be determined that it is a living body only when both the left and right eyes are determined to be living bodies, or it may be determined that it is a living body when only one of the left and right eyes is determined to be a living body.

[0045] <Summary> The information processing terminal 1 illuminates the eyes of the subject 2, which is the object of living body determination, with one infrared light source 10, and captures images of the eyes of the subject 2 with the infrared camera 13 to obtain infrared images of the eyes of the subject 2, and obtains visible light images of the eyes of the subject 2 with the visible light camera 12.

[0046] The information processing terminal 1 determines whether there is a corneal reflex in the infrared image of the eyes of the subject 2 (this determination may be referred to as the "first corneal reflex determination"), and determines whether there is pupil absorption in the infrared image of the eyes of the subject 2 (this determination may be referred to as the "first pupil absorption determination").

[0047] The information processing terminal 1 determines whether there is a corneal reflex in the visible light image of the eyes of the subject 2 (this determination may be referred to as the "second corneal reflex determination"), and determines whether there is pupil absorption in the visible light image of the eyes of the subject 2 (this determination may be referred to as the "second pupil absorption determination").

[0048] The information processing terminal 1 determines whether the subject 2 is a living body by determining whether a living body determination condition, which is a condition indicating that the subject 2 is a living body, is satisfied based on these above determination results.

[0049] The biometric determination condition is defined based on the characteristics of the infrared image and the visible light image when the subject 2 is a living body, as described below. The characteristics of the infrared image and the visible light image when the subject 2 is the living body (person 3000) shown in FIG. 3C will be described. FIG. 3A is a diagram for explaining the characteristics of the infrared image of the eye of the person 3000. FIG. 3A includes the infrared image of the eye of the person 3000, a schematic diagram showing the state of the eye of the person 3000, and a histogram of the pixel values of the inscribed rectangular area of the iris area of the infrared image of the eye of the person 3000.

[0050] As shown in FIG. 3A, there is one corneal reflection in the infrared image of the eye of the person 3000. Further, in the histogram of the infrared image of the eye of the person 3000, there are pixels having a relatively bright pixel value indicating one corneal reflection, and pixels having a relatively dark pixel value indicating pupil absorption.

[0051] FIG. 3B is a diagram for explaining the characteristics of the visible light image of the eye of the person 3000. FIG. 3B includes the visible light image of the eye of the person 3000, a schematic diagram showing the state of the eye of the person 3000, and a histogram of the pixel values of the inscribed rectangular area of the iris area of the visible light image of the eye of the person 3000.

[0052] As shown in FIG. 3B, there is no corneal reflection in the visible light image of the eye of the person 3000. Further, in the histogram of the visible light image of the eye of the person 3000, there are no pixels having a relatively bright pixel value indicating corneal reflection, and there are pixels having a relatively dark pixel value indicating pupil absorption.

[0053] In view of the above, when there is a corneal reflection in the infrared image of the eye of the subject 2, and there is pupil absorption, and there is no corneal reflection in the visible light image of the eye of the subject 2, and there is pupil absorption, the possibility of being a living body becomes quite high. Therefore, it can be defined that there is a corneal reflection in the infrared image of the eye of the subject 2, and there is pupil absorption, and there is no corneal reflection in the visible light image of the eye of the subject 2, and there is pupil absorption as the biometric determination condition.

[0054] The information processing terminal 1 uses as a biological determination condition that there is a corneal reflex in the infrared image of the eye of the subject 2, and there is pupil absorption, and there is no corneal reflex in the visible light image of the eye of the subject 2, and there is pupil absorption. When the biological determination condition is satisfied, it determines that the subject 2 is a living body, and when the biological determination condition is not satisfied, it determines that the subject 2 is not a living body.

[0055] By thus determining whether the subject 2 is a living body or not, the information processing terminal 1 can correctly determine that the subject 2 is a living body when the subject 2 is a living body such as the person 3000. Also, when the subject 2 is a fake image of the person 3000 such as the printed matter (1) 3001 and the printed matter (2) 3002 shown below, the information processing terminal 1 can correctly determine that the subject 2 is not a living body.

[0056] FIG. 4A is a diagram for explaining the characteristics of the infrared image of the eye of the printed matter (1) 3001. FIG. 4A includes the infrared image of the eye of the printed matter (1) 3001, a schematic diagram showing the state of the eye of the printed matter (1) 3001, and a histogram of the pixel values of the inscribed rectangular region of the iris region of the infrared image of the eye of the printed matter (1) 3001. The printed matter (1) 3001 is a fake image of a person. Note that there is no corneal reflex image in the eye of the fake image.

[0057] There is no corneal reflex in the infrared image of the eye of the printed matter (1) 3001. Further, in the histogram of the infrared image of the eye of the printed matter (1) 3001, there are no pixels having a relatively bright pixel value indicating a corneal reflex. In the histogram of the infrared image of the eye of the printed matter (1) 3001, there are relatively dark pixels indicating pupil absorption. Note that depending on the imaging situation of the printed matter (1) 3001, since the light of the infrared light source 10 is reflected by the black portion which is a fake of the pupil and a certain amount of light returns to the camera side, the histogram of the infrared image of the eye of the printed matter (1) 3001 may be such that there are no relatively dark pixels indicating pupil absorption as shown in the histogram of FIG. 4D.

[0058] FIG. 4B is a diagram for explaining the characteristics of the visible light image of the eye of the printed matter (1) 3001. FIG. 4B includes the visible light image of the eye of the printed matter (1) 3001, a schematic diagram showing the state of the eye of the printed matter (1) 3001, and a histogram of the pixel values of the inscribed rectangular area of the iris area of the visible light image of the eye of the printed matter (1) 3001.

[0059] There is no corneal reflex in the visible light image of the eye of the printed matter (1) 3001. Furthermore, in the histogram of the visible light image of the eye of the printed matter (1) 3001, there are pixels having relatively dark pixel values indicating pupil absorption. Note that depending on the imaging situation of the printed matter (1) 3001, since visible light is reflected in the black part which is a fake of the pupil and a certain amount of light returns to the camera side, the histogram of the visible light image of the eye of the printed matter (1) 3001 may not have relatively dark pixels indicating pupil absorption as in the histogram shown in FIG. 4D.

[0060] When the subject 2 is the printed matter (1) 3001, since there is no corneal reflex in the infrared image of the eye, the living body determination condition is not satisfied. Therefore, when the subject 2 is the printed matter (1) 3001, the information processing terminal 1 can correctly determine that the subject 2 is not a living body.

[0061] FIG. 5A is a diagram for explaining the characteristics of the infrared image of the eye of the printed matter (2) 3002. FIG. 5A includes the infrared image of the eye of the printed matter (2) 3002, a schematic diagram showing the state of the eye of the printed matter (2) 3002, and a histogram of the pixel values of the inscribed rectangular area of the iris area of the infrared image of the eye of the printed matter (2) 3002. The printed matter (2) 3002 is a fake image of a person, and is different from the printed matter (1) 3001 in that there is a corneal reflection image in the eye of the person in the fake image.

[0062] There is one corneal reflex in the infrared image of the eye of the printed matter (2) 3002. Furthermore, in the histogram of the infrared image of the eye of the printed matter (2) 3002, there are pixels having relatively bright pixel values indicating one corneal reflex and pixels having relatively dark pixel values indicating pupil absorption.

[0063] FIG. 5B is a diagram for explaining the characteristics of the visible light image of the eye of the printed matter (2) 3002. FIG. 5B includes a visible light image of the eye of the printed matter (2) 3002, a schematic diagram showing the state of the eye of the printed matter (2) 3002, and a histogram of the pixel values of the inscribed rectangular area of the iris area of the visible light image of the eye of the printed matter (2) 3002.

[0064] In the visible light image of the eye of the printed matter (2) 3002, there is one corneal reflection. Furthermore, in the histogram of the visible light image of the eye of the printed matter (2) 3002, there are pixels having a relatively bright pixel value indicating one corneal reflection and pixels having a relatively dark pixel value indicating pupil absorption.

[0065] When the subject 2 is the printed matter (2) 3002, since there is a corneal reflection in the visible light image of the eye, the living body determination condition is not satisfied. Therefore, the information processing terminal 1 can correctly determine that the subject 2 is not a living body even if the subject 2 is a fake image of a person with a corneal reflection image in the eye like the printed matter (2) 3002.

[0066] <Specific operation> FIG. 6 is a flowchart showing the processing flow executed by the living body determination unit 105. The living body determination unit 105 starts the processing from step 600 and proceeds to step 601, and the pupil absorption determination unit 103 determines whether there is pupil absorption in the infrared image of the eye of the subject 2 that is the target of living body determination.

[0067] If there is no pupil absorption in the infrared image of the eye of the subject 2, it is highly likely that the subject 2 is not a living body. Therefore, in this case, the living body determination unit 105 determines "No" in step 601, proceeds to step 602, determines that the subject 2 is not a living body, proceeds to step 695, and temporarily ends this processing flow.

[0068] When there is pupil absorption in the infrared image of the eyes of the subject 2, in this determination, it is highly likely that the subject 2 is a living body (the possibility of not being a living body is low). Therefore, in this case, the living body determination unit 105 determines "Yes" in step 601 and proceeds to step 603, and determines whether or not there is pupil absorption in the visible light image of the eyes of the subject 2 by the pupil absorption determination unit 103.

[0069] When there is no pupil absorption in the visible light image of the eyes of the subject 2, it is highly likely that the subject 2 is not a living body. Therefore, in this case, the living body determination unit 105 determines "No" in step 603 and proceeds to step 602, determines that the subject 2 is not a living body, proceeds to step 695, and once ends this processing flow.

[0070] When there is pupil absorption in the visible light image of the eyes of the subject 2, in this determination, it is highly likely that the subject 2 is a living body (the possibility of not being a living body is low). Therefore, in this case, the living body determination unit 105 determines "Yes" in step 603 and proceeds to step 604, and determines whether or not there is corneal reflection in the infrared image of the eyes of the subject 2 by the corneal reflection determination unit 104.

[0071] When there is no corneal reflection in the infrared image of the eyes of the subject 2, it is highly likely that the subject 2 is not a living body. Therefore, in this case, the living body determination unit 105 determines "No" in step 604 and proceeds to step 602, determines that the subject 2 is not a living body, proceeds to step 695, and once ends this processing flow.

[0072] When there is corneal reflection in the infrared image of the eyes of the subject 2, in this determination, it is highly likely that the subject 2 is a living body (the possibility of not being a living body is low). Therefore, in this case, the living body determination unit 105 determines "Yes" in step 604 and proceeds to step 605, and determines whether or not there is no corneal reflection in the visible light image of the eyes of the subject 2 by the corneal reflection determination unit 104.

[0073] If there is a corneal reflex in the visible light image of the eyes of the subject 2, it is highly likely that the subject 2 is not a living body. Therefore, in this case, the living body determination unit 105 determines "No" in step 605, proceeds to step 602, determines that the subject 2 is not a living body, proceeds to step 695, and once ends this processing flow.

[0074] If there is no corneal reflex in the visible light image of the eyes of the subject 2, in this determination, it is highly likely that the subject 2 is a living body (the possibility of not being a living body is low). Therefore, in this case, the living body determination unit 105 determines "Yes" in step 605 and proceeds to step 606.

[0075] When the living body determination unit 105 proceeds to step 606, there is pupil absorption in the infrared image of the eyes of the subject 2 ( "Yes" in step 601), and there is pupil absorption in the visible light image of the eyes of the subject 2 ( "Yes" in step 603), and there is a corneal reflex in the infrared image of the eyes of the subject 2 ( "Yes" in step 604), and there is no corneal reflex in the visible light image of the eyes of the subject 2 ( "Yes" in step 605). Since the living body determination conditions are satisfied, the possibility that the subject 2 is a living body becomes quite high. Therefore, the living body determination unit 105 determines in step 606 that the subject 2 is a living body, proceeds to step 695, and once ends this processing flow.

[0076] <Effect> As described above, the information processing terminal 1 according to the first embodiment of the present invention can accurately determine whether the subject 2 is a living body. When the subject 2 is a fake image (printed matter (1) 3001) of a person 3000 without a corneal reflex image in the eyes as shown in FIG. 4C, the information processing terminal 1 according to the first embodiment can accurately determine that the subject 2 is not a living body. Even when the subject 2 is a fake image (printed matter (2) 3002) of a person 3000 with a corneal reflex image in the eyes as shown in FIG. 5C, the information processing terminal 1 according to the first embodiment can accurately determine that the subject 2 is not a living body.

[0077] <<Second Embodiment>> The information processing terminal 1 according to the second embodiment of the present invention will be described. The information processing terminal 1 according to the second embodiment has differences from the information processing terminal 1 according to the first embodiment only in the following points. · The information processing terminal 1 according to the second embodiment has two infrared light sources 10, and the biometric determination conditions are different from those of the first embodiment.

[0078] Hereinafter, the description will focus on this difference.

[0079] <Configuration> FIG. 7 is a configuration diagram showing a hardware configuration example of the information processing terminal 1 according to the second embodiment. FIG. 8 is a diagram showing a functional block of the controller 11 of the information processing terminal 1 in FIG. 7. As shown in FIGS. 7 and 8, the information processing terminal 1 according to the second embodiment has two infrared light sources 10.

[0080] <Overview> The information processing terminal 1 illuminates the eyes of the subject 2, which is the biometric determination target, with two infrared light sources 10, and captures an infrared image of the eyes of the subject 2 with the infrared camera 13 while illuminating the eyes of the subject 2, and acquires a visible light image of the eyes of the subject 2 with the visible light camera 12.

[0081] The information processing terminal 1 determines the number of corneal reflections when there is a corneal reflection in the infrared image of the eyes of the subject 2 (this determination may also be referred to as the "first corneal reflection determination"), and determines whether there is pupil absorption in the infrared image of the eyes of the subject 2 (performs the "first pupil absorption determination").

[0082] The information processing terminal 1 determines whether there is a corneal reflection in the visible light image of the eyes of the subject 2 (performs the "second corneal reflection determination"), and determines whether there is pupil absorption in the visible light image of the eyes of the subject 2 (performs the "second pupil absorption determination").

[0083] The information processing terminal 1 determines whether the biometric determination conditions are satisfied based on these determination results described above, and thereby determines whether the subject 2 is a living body.

[0084] The biometric determination conditions are defined based on the characteristics of the infrared image and the visible light image when the subject 2 is a living body, as described below.

[0085] The characteristics of the infrared image and the visible light image when the subject 2 is the living body (person 4000) shown in FIG. 9C will be described. FIG. 9A is a diagram for explaining the characteristics of the infrared image of the eye of the person 4000. FIG. 9A includes the infrared image of the eye of the person 4000, a schematic diagram showing the state of the eye of the person 4000, and a histogram of the pixel values of the inscribed rectangular region of the iris region of the infrared image of the eye of the person 4000.

[0086] As shown in FIG. 9A, there are two corneal reflections in the infrared image of the eye of the person 4000. Further, in the histogram of the infrared image of the eye of the person 4000, there are pixels having relatively bright pixel values indicating the two corneal reflections, and there are pixels having relatively dark pixel values indicating pupil absorption.

[0087] FIG. 9B is a diagram for explaining the characteristics of the visible light image of the eye of the person 4000. FIG. 9B includes the visible light image of the eye of the person 4000, a schematic diagram showing the state of the eye of the person 4000, and a histogram of the pixel values of the inscribed rectangular region of the iris region of the visible light image of the eye of the person 4000.

[0088] As shown in FIG. 9B, there is no corneal reflection in the visible light image of the eye of the person 4000. Further, in the histogram of the visible light image of the eye of the person 4000, there are no pixels having relatively bright pixel values indicating corneal reflection, and there are pixels having relatively dark pixel values indicating pupil absorption.

[0089] In view of the above, when there are two corneal reflections in the infrared image of the eye of the subject 2 (the same number as the infrared light source 10), and there is pupil absorption, and there is no corneal reflection in the visible light image of the eye of the subject 2, and there is pupil absorption, the possibility that the subject 2 is a living body becomes quite high. Therefore, it can be stipulated that having two corneal reflections in the infrared image of the eye of the subject 2 (the same number as the infrared light source 10), and having pupil absorption, and having no corneal reflection in the visible light image of the eye of the subject 2, and having pupil absorption is used as the biological determination condition.

[0090] The information processing terminal 1 uses as the biological determination condition that there are two corneal reflections in the infrared image of the eye of the subject 2 (the same number as the infrared light source 10), and there is pupil absorption, and there is no corneal reflection in the visible light image of the eye of the subject 2, and there is pupil absorption. When the biological determination condition is satisfied, it determines that the subject 2 is a living body, and when the biological determination condition is not satisfied, it determines that the subject 2 is not a living body.

[0091] By thus determining whether the subject 2 is a living body or not, the information processing terminal 1 can correctly determine that the subject 2 is a living body when the subject 2 is the person 4000. In addition, when the subject 2 is a fake image of a person such as the printed matter (1) 4001 and the printed matter (2) 4002 shown below, the information processing terminal 1 can correctly determine that the subject 2 is not a living body.

[0092] FIG. 10A is a diagram for explaining the characteristics of the infrared image of the eye of the printed matter (1) 4001. FIG. 10A includes the infrared image of the eye of the printed matter (1) 4001, a schematic diagram showing the state of the eye of the printed matter (1) 4001, and a histogram of the pixel values of the inscribed rectangular area of the iris area of the infrared image of the eye of the printed matter (1) 4001. The printed matter (1) 4001 is a fake image of a person. Note that there is no corneal reflection image in the eye of the fake image.

[0093] In the infrared image of the eye of the printed matter (1) 4001, there is no corneal reflex. Furthermore, in the histogram of the infrared image of the eye of the printed matter (1) 4001, there are no pixels having a relatively bright pixel value indicating a corneal reflex. In the histogram of the infrared image of the eye of the printed matter (1) 4001, there are relatively dark pixels indicating pupil absorption.

[0094] FIG. 10B is a diagram for explaining the characteristics of the visible light image of the eye of the printed matter (1) 4001. FIG. 10B includes a visible light image of the eye of the printed matter (1) 4001, a schematic diagram showing the state of the eye of the printed matter (1) 4001, and a histogram of pixel values of an inscribed rectangular area of the iris area of the visible light image of the eye of the printed matter (1) 4001.

[0095] In the visible light image of the eye of the printed matter (1) 4001, there is no corneal reflex. Furthermore, in the histogram of the visible light image of the eye of the printed matter (1) 4001, there are pixels having a relatively dark pixel value indicating pupil absorption.

[0096] When the subject 2 is the printed matter (1) 4001, since there are no two corneal reflexes in the infrared image of the eye, the biometric determination condition is not satisfied. Therefore, when the subject 2 is the printed matter (1) 4001, the information processing terminal 1 can correctly determine that the subject 2 is not a living body.

[0097] FIG. 11A is a diagram for explaining the characteristics of the infrared image of the eye of the printed matter (2) 4002. FIG. 11A includes an infrared image of the eye of the printed matter (2) 4002, a schematic diagram showing the state of the eye of the printed matter (2) 4002, and a histogram of pixel values of an inscribed rectangular area of the iris area of the infrared image of the eye of the printed matter (2) 4002. The printed matter (2) 4002 is a fake image of a person, and is different from the printed matter (1) 4001 in that there is one corneal reflex image in the eye of the person in the fake image.

[0098] In the infrared image of the eye of the printed matter (2) 4002, there is one corneal reflex. Furthermore, in the histogram of the infrared image of the eye of the printed matter (2) 4002, there are pixels having a relatively bright pixel value indicating one corneal reflex, and there are pixels having a relatively dark pixel value indicating pupil absorption.

[0099] FIG. 11B is a diagram for explaining the characteristics of the visible light image of the eye of the printed matter (2) 4002. FIG. 11B includes a visible light image of the eye of the printed matter (2) 4002, a schematic diagram showing the state of the eye of the printed matter (2) 4002, and a histogram of the pixel values of the inscribed rectangular area of the iris area of the visible light image of the eye of the printed matter (2) 4002.

[0100] In the visible light image of the eye of the printed matter (2) 4002, there is one corneal reflex. Further, in the histogram of the visible light image of the eye of the printed matter (2) 4002, there are pixels having a relatively bright pixel value indicating one corneal reflex and pixels having a relatively dark pixel value indicating pupil absorption.

[0101] When the subject 2 is the printed matter (2) 4002, there is only one corneal reflex in the infrared image of the eye, and there is a corneal reflex in the visible light image of the eye, so the biometric determination condition is not satisfied. Therefore, the information processing terminal 1 can correctly determine that the subject 2 is not a living body even if the subject 2 is a fake image of a person with a corneal reflex image in the eye like the printed matter (2) 4002.

[0102] <Specific operation> FIG. 12 is a flowchart showing the processing flow executed by the biometric determination unit 105. The biometric determination unit 105 starts processing from step 1200 and proceeds to step 1201, where the pupil absorption determination unit 103 determines whether there is pupil absorption in the infrared image of the eye of the subject 2 that is the object of biometric determination.

[0103] If there is no pupil absorption in the infrared image of the eye of the subject 2, it is highly likely that the subject 2 is not a living body. Therefore, in this case, the biometric determination unit 105 determines "No" in step 1201, proceeds to step 1202, determines that the subject 2 is not a living body, proceeds to step 1295, and temporarily ends this processing flow.

[0104] When there is pupil absorption in the infrared image of the eyes of the subject 2, in this determination, the possibility that the subject 2 is a living body is high (the possibility of not being a living body is low). Therefore, in this case, the living body determination unit 105 determines "Yes" in step 1201 and proceeds to step 1203, and determines whether or not it is determined that there is pupil absorption in the visible light image of the eyes of the subject 2 by the pupil absorption determination unit 103.

[0105] When there is no pupil absorption in the visible light image of the eyes of the subject 2, the possibility that the subject 2 is not a living body is high. Therefore, in this case, the living body determination unit 105 determines "No" in step 1203 and proceeds to step 1202, determines that the subject 2 is not a living body, proceeds to step 1295, and once ends this processing flow.

[0106] When there is pupil absorption in the visible light image of the eyes of the subject 2, in this determination, the possibility that the subject 2 is a living body is high (the possibility of not being a living body is low). Therefore, in this case, the living body determination unit 105 determines "Yes" in step 1203 and proceeds to step 1204, and determines whether or not it is determined that there are two corneal reflections in the infrared image of the eyes of the subject 2 by the corneal reflection determination unit 104.

[0107] When there are not two corneal reflections in the infrared image of the eyes of the subject 2, the possibility that the subject 2 is not a living body is high. Therefore, in this case, the living body determination unit 105 determines "No" in step 1204 and proceeds to step 1202, determines that the subject 2 is not a living body, proceeds to step 1295, and once ends this processing flow.

[0108] When there are two corneal reflections in the infrared image of the eyes of the subject 2, in this determination, the possibility that the subject 2 is a living body is high (the possibility of not being a living body is low). Therefore, in this case, the living body determination unit 105 determines "Yes" in step 1204 and proceeds to step 1205, and determines whether or not there is no corneal reflection in the visible light image of the eyes of the subject 2 by the corneal reflection determination unit 104.

[0109] If there is a corneal reflex in the visible light image of the eyes of the subject 2, it is highly likely that the subject 2 is not a living body. Therefore, in this case, the living body determination unit 105 determines "No" in step 1205, proceeds to step 1202, determines that the subject 2 is not a living body, proceeds to step 1295, and temporarily ends this processing flow.

[0110] If there is no corneal reflex in the visible light image of the eyes of the subject 2, it is highly likely (low possibility of not being a living body) that the subject 2 is a living body in this determination. Therefore, in this case, the living body determination unit 105 determines "Yes" in step 1205 and proceeds to step 1206.

[0111] When the living body determination unit 105 proceeds to step 1206, there is pupil absorption in the infrared image of the eyes of the subject 2 ( "Yes" in step 1201), there is pupil absorption in the visible light image of the eyes of the subject 2 ( "Yes" in step 1203), there are two corneal reflexes in the infrared image of the eyes of the subject 2 ( "Yes" in step 1204), and there is no corneal reflex in the visible light image of the eyes of the subject 2 ( "Yes" in step 1205). Since the living body determination conditions are satisfied, the possibility that the subject 2 is a living body becomes quite high. Therefore, the living body determination unit 105 determines in step 1206 that the subject 2 is a living body, proceeds to step 1295, and temporarily ends this processing flow.

[0112] <Effect> As described above, the information processing terminal 1 according to the second embodiment of the present invention can accurately determine whether the subject 2 is a living body. When the subject 2 is a fake image (printed matter (1) 4001) of a person without a corneal reflex image in the eyes as shown in FIG. 10C, the information processing terminal 1 according to the second embodiment can accurately determine that the subject 2 is not a living body. Even when the subject 2 is a fake image (printed matter (2) 4002) of a person with a corneal reflex image in the eyes as shown in FIG. 11C, the information processing terminal 1 according to the second embodiment can accurately determine that the subject 2 is not a living body.

[0113] <<Third Embodiment>> The information processing terminal 1 according to the third embodiment of the present invention will be described. The information processing terminal 1 according to the third embodiment has differences from the information processing terminal 1 according to the second embodiment only in the following points. · The information processing terminal 1 according to the third embodiment changes the number of light emissions (lighting numbers) of the infrared light source 10, and determines whether there are corneal reflections in the infrared image of the eye of the subject 2 in a number (the same number as the number of light emissions) corresponding to the number of light emissions (lighting numbers). The information processing terminal 1 according to the third embodiment adopts a biometric determination condition different from the biometric determination condition of the second embodiment as the biometric determination condition.

[0114] Hereinafter, the description will be centered on this difference.

[0115] <Overview> When the subject 2 is a living body, two corneal reflections exist in the infrared image of the eye of the subject 2 imaged in a state where the eye of the subject 2 is illuminated by two infrared light sources 10. When the subject 2 is a living body, one corneal reflection exists in the infrared image of the eye of the subject 2 imaged in a state where the eye of the subject 2 is illuminated by one infrared light source 10. That is, when the subject 2 is a living body, if corneal reflections (the same number of corneal reflections as the number of light emissions of the infrared light source 10) corresponding to the number of light emissions (lighting numbers) of the infrared light source 10 exist in the infrared image of the eye of the subject 2, the possibility that the subject 2 is a living body becomes quite high.

[0116] Therefore, the information processing terminal 1 according to the third embodiment adopts, as the biometric determination condition, that there is pupil absorption in the infrared image of the eye of the subject 2, there is pupil absorption in the visible light image of the eye of the subject 2, there are two corneal reflections in the infrared image of the eye of the subject 2 imaged in a state where the eye of the subject 2 is illuminated by two infrared light sources 10, there is one corneal reflection in the infrared image of the eye of the subject 2 imaged in a state where the eye of the subject 2 is illuminated by one infrared light source 10, and there is no corneal reflection in the visible light image of the eye of the subject 2 imaged in a state where the eye of the subject 2 is illuminated by the infrared light source 10 (one or two infrared light sources 10).

[0117] <Specific operation> FIG. 13 is a flowchart showing the processing flow executed by the biological determination unit 105. The biological determination unit 105 starts processing from step 1300 and proceeds to step 1301, where the pupil absorption determination unit 103 determines whether there is pupil absorption in the infrared image of the eye of the subject 2.

[0118] If there is no pupil absorption in the infrared image of the eye of the subject 2, it is highly likely that the subject 2 is not a living body. Therefore, in this case, the biological determination unit 105 determines "No" in step 1301, proceeds to step 1302, determines that the subject 2 is not a living body, proceeds to step 1395, and temporarily ends this processing flow.

[0119] If there is pupil absorption in the infrared image of the eye of the subject 2, in this determination, it is highly likely that the subject 2 is a living body (the possibility of not being a living body is low). Therefore, in this case, the biological determination unit 105 determines "Yes" in step 1301, proceeds to step 1303, and the pupil absorption determination unit 103 determines whether it is determined that there is pupil absorption in the visible light image of the eye of the subject 2.

[0120] If there is no pupil absorption in the visible light image of the eye of the subject 2, it is highly likely that the subject 2 is not a living body. Therefore, in this case, the biological determination unit 105 determines "No" in step 1303, proceeds to step 1302, determines that the subject 2 is not a living body, proceeds to step 1395, and temporarily ends this processing flow.

[0121] If there is pupil absorption in the visible light image of the eye of the subject 2, in this determination, it is highly likely that the subject 2 is a living body (the possibility of not being a living body is low). Therefore, in this case, the biological determination unit 105 determines "Yes" in step 1303, proceeds to step 1304, and the light source control unit 115 drives two infrared light sources 10 and proceeds to step 1305.

[0122] When the biological determination unit 105 proceeds to step 1305, the corneal reflection determination unit 104 determines whether there are two corneal reflections in the infrared image of the eye of the subject 2.

[0123] If there are no two corneal reflections in the infrared image of the eyes of the subject 2, it is highly likely that the subject 2 is not a living body. Therefore, in this case, the living body determination unit 105 determines "No" in step 1305, proceeds to step 1302, determines that the subject 2 is not a living body, proceeds to step 1395, and once ends this processing flow.

[0124] If there are two corneal reflections in the infrared image of the eyes of the subject 2, in this determination, it is highly likely that the subject 2 is a living body (the possibility of not being a living body is low). Therefore, in this case, the living body determination unit 105 determines "Yes" in step 1305, proceeds to step 1306, drives one infrared light source 10, and proceeds to step 1307.

[0125] When proceeding to step 1307, the living body determination unit 105 determines whether there is one corneal reflection in the infrared image of the eyes of the subject 2 by the corneal reflection determination unit 104.

[0126] If there is no corneal reflection in the infrared image of the eyes of the subject 2, it is highly likely that the subject 2 is not a living body. Therefore, in this case, the living body determination unit 105 determines "No" in step 1307, proceeds to step 1302, determines that the subject 2 is not a living body, proceeds to step 1395, and once ends this processing flow.

[0127] If there is one corneal reflection in the infrared image of the eyes of the subject 2, in this determination, it is highly likely that the subject 2 is a living body (the possibility of not being a living body is low). Therefore, in this case, the living body determination unit 105 determines "Yes" in step 1307, proceeds to step 1308, and determines whether there is no corneal reflection in the visible light image of the eyes of the subject 2 by the corneal reflection determination unit 104.

[0128] If there is a corneal reflection in the visible light image of the eyes of the subject 2, it is highly likely that the subject 2 is not a living body. Therefore, in this case, the living body determination unit 105 determines "No" in step 1308, proceeds to step 1302, determines that the subject 2 is not a living body, proceeds to step 1395, and once ends this processing flow.

[0129] When there is no corneal reflex in the visible light image of the eyes of the subject 2, in this determination, it is highly likely that the subject 2 is a living body (the possibility that the subject 2 is not a living body is low). Therefore, in this case, the living body determination unit 105 determines "Yes" in step 1308 and proceeds to step 1309.

[0130] When the living body determination unit 105 proceeds to step 1309, there is pupil absorption in the infrared image of the eyes of the subject 2 ( "Yes" in step 1301), and there is pupil absorption in the visible light image of the eyes of the subject 2 ( "Yes" in step 1303), and there are two corneal reflexes in the infrared image of the eyes of the subject 2 captured in a state where the eyes of the subject 2 are illuminated by two infrared light sources 10 ( "Yes" in step 1305), and there is one corneal reflex in the infrared image of the eyes of the subject 2 captured in a state where the eyes of the subject 2 are illuminated by one infrared light source 10 ( "Yes" in step 1307), and there is no corneal reflex in the visible light image of the eyes of the subject 2 captured in a state where the eyes of the subject 2 are illuminated by one infrared light source 10 ( "Yes" in step 1308). Since the living body determination conditions are satisfied, the possibility that the subject 2 is a living body becomes quite high. Therefore, the living body determination unit 105 determines in step 1309 that the subject 2 is a living body, proceeds to step 1395, and once ends this processing flow.

[0131] <Effect> As described above, the information processing terminal 1 according to the third embodiment of the present invention can accurately determine whether the subject 2 is a living body. The information processing terminal 1 according to the third embodiment can accurately determine that the subject 2 is not a living body when the subject 2 is a fake image (printed matter (1) 4001) of a person without a corneal reflex image in the eyes as shown in FIG. 10C. The information processing terminal 1 according to the third embodiment can accurately determine that the subject 2 is not a living body even when the subject 2 is a fake image (printed matter (2) 4002) of a person with a corneal reflex image in the eyes as shown in FIG. 11C.

[0132] <<Fourth Embodiment>> The information processing terminal 1 according to the fourth embodiment of the present invention will be described. The information processing terminal 1 according to the fourth embodiment has differences from the information processing terminal 1 according to the first embodiment only in the following points. · The information processing terminal 1 according to the fourth embodiment has two infrared light sources 10 and two visible light sources 16, and determines whether there are the same number of corneal reflections (the same number as the number of light emissions) as the number of light emissions (the number of lit lights) in the infrared image and the visible light image of the eyes of the subject 2. The information processing terminal 1 according to the fourth embodiment adopts a biometric determination condition different from the biometric determination condition of the first embodiment as the biometric determination condition.

[0133] Hereinafter, the description will focus on this difference.

[0134] <Configuration> FIG. 14 is a diagram showing a functional block of the controller 11 of the information processing terminal 1 according to the fourth embodiment. As shown in FIG. 14, the information processing terminal 1 according to the fourth embodiment has two infrared light sources 10 and two visible light sources 16. From the viewpoint of realizing a simpler configuration in the information processing terminal 1, it is preferable that the optical axis of the visible light camera 12 and the optical axes of the two visible light sources 16 are different from each other. From the viewpoint of realizing a simpler configuration in the information processing terminal 1, it is preferable that the optical axis of the infrared camera 13 and the optical axes of the two infrared light sources 10 are different from each other. From the viewpoint of realizing a simpler configuration in the information processing terminal 1, it is preferable that the optical axis of the visible light camera 12 and the optical axes of the two infrared light sources 10 are different from each other. From the viewpoint of realizing a simpler configuration in the information processing terminal 1, it is preferable that the optical axis of the infrared camera 13 and the optical axes of the two visible light sources 16 are different from each other.

[0135] <Overview> The information processing terminal 1 illuminates the eyes of the subject 2 with two infrared light sources 10, and captures an infrared image of the eyes of the subject 2, which is the object of biometric determination, by using the infrared camera 13, and obtains a visible light image of the eyes of the subject 2, which is the object of biometric determination, by using the visible light camera 12 in a state where the eyes of the subject 2 are irradiated with two visible light sources 16.

[0136] The information processing terminal 1 determines the number of corneal reflections when there are corneal reflections in the infrared image of the eyes of the subject 2 (performs "first corneal reflection determination"), and determines whether there is pupil absorption in the infrared image of the eyes of the subject 2 (performs "first pupil absorption determination").

[0137] The information processing terminal 1 determines the number of corneal reflections when there are corneal reflections in the visible light image of the eyes of the subject 2 (this determination may also be referred to as "second corneal reflection determination"), and determines whether there is pupil absorption in the visible light image of the eyes of the subject 2 (performs "second pupil absorption determination").

[0138] Based on these determination results described above, the information processing terminal 1 determines whether the biometric determination condition is satisfied, thereby determining whether the subject 2 is a living body. The biometric determination condition is defined based on the characteristics of the infrared image and the visible light image when the subject 2 is a living body, as described below.

[0139] The characteristics of the infrared image and the visible light image when the subject 2 is the living body (person 5000) shown in FIG. 15C will be described. FIG. 15A is a diagram for explaining the characteristics of the infrared image of the eyes of the person 5000. FIG. 15A includes the infrared image of the eyes of the person 5000, a schematic diagram showing the state of the eyes of the person 5000, and a histogram of the pixel values of the inscribed rectangular region of the iris region of the infrared image of the eyes of the person 5000.

[0140] As shown in FIG. 15A, there are two corneal reflections in the infrared image of the eyes of the person 5000. Further, in the histogram of the infrared image of the eyes of the person 5000, there are pixels having relatively bright pixel values indicating two corneal reflections, and pixels having relatively dark pixel values indicating pupil absorption.

[0141] FIG. 15B is a diagram for explaining the characteristics of the visible light image of the eye of person 5000. FIG. 15B includes a schematic diagram showing the state of the eye of person 5000 and a histogram of the pixel values of the inscribed rectangular area of the iris region of the visible light image of the eye of person 5000.

[0142] As shown in FIG. 15B, there are two corneal reflections in the visible light image of the eye of person 5000. Furthermore, in the histogram of the visible light image of the eye of person 5000, there are pixels having relatively bright pixel values indicating two corneal reflections and pixels having relatively dark pixel values indicating pupil absorption.

[0143] In view of the above, when there are two corneal reflections in the infrared image of the eye of subject 2 (the same number as that of the infrared light source 10), and there is pupil absorption, and there are two corneal reflections in the visible light image of the eye of subject 2 (the same number as that of the visible light source 16), and there is pupil absorption, the possibility that subject 2 is a living body becomes quite high. Therefore, it can be defined that having two corneal reflections in the infrared image of the eye of subject 2 (the same number as that of the infrared light source 10), and having pupil absorption, and having two corneal reflections in the visible light image of the eye of subject 2 (the same number as that of the visible light source 16), and having pupil absorption is used as a biological determination condition.

[0144] The information processing terminal 1 uses as a biological determination condition that there are two corneal reflections in the infrared image of the eye of subject 2 (the same number as that of the infrared light source 10), and there is pupil absorption, and there are two corneal reflections in the visible light image of the eye of subject 2 (the same number as that of the visible light source 16), and there is pupil absorption. When the biological determination condition is satisfied, it determines that subject 2 is a living body, and when the biological determination condition is not satisfied, it determines that subject 2 is not a living body.

[0145] By determining whether or not the subject 2 is a living body in this way, the information processing terminal 1 can correctly determine that the subject 2 is a living body when the subject 2 is a person 5000. Further, when the subject 2 is a fake image of a person such as the printed matter (1) 5001 and the printed matter (2) 5002 shown below, the information processing terminal 1 can correctly determine that the subject 2 is not a living body.

[0146] FIG. 16A is a diagram for explaining the characteristics of the infrared image of the eye of the printed matter (1) 5001. FIG. 16A includes the infrared image of the eye of the printed matter (1) 5001, a schematic diagram showing the state of the eye of the printed matter (1) 5001, and a histogram of the pixel values of the inscribed rectangular area of the iris area of the infrared image of the eye of the printed matter (1) 5001. The printed matter (1) 5001 is a fake image of a person. Note that there is no corneal reflection image in the eye of the fake image.

[0147] There is no corneal reflection in the infrared image of the eye of the printed matter (1) 5001. Further, in the histogram of the infrared image of the eye of the printed matter (1) 5001, there are no pixels having a relatively bright pixel value indicating corneal reflection. In the histogram of the infrared image of the eye of the printed matter (1) 5001, there are pixels having a relatively dark pixel value indicating pupil absorption.

[0148] FIG. 16B is a diagram for explaining the characteristics of the visible light image of the eye of the printed matter (1) 5001. FIG. 16B includes the visible light image of the eye of the printed matter (1) 5001, a schematic diagram showing the state of the eye of the printed matter (1) 5001, and a histogram of the pixel values of the inscribed rectangular area of the iris area of the visible light image of the eye of the printed matter (1) 5001.

[0149] There is no corneal reflection in the visible light image of the eye of the printed matter (1) 5001. In the histogram of the visible light image of the eye of the printed matter (1) 5001, there are pixels having a relatively dark pixel value indicating pupil absorption.

[0150] When the object 2 is the printed matter (1) 5001, there are no two corneal reflections in the infrared image of the eye, and there are no two corneal reflections in the visible light image of the eye, so the biometric determination condition is not satisfied. Therefore, when the object 2 is the printed matter (1) 5001, the information processing terminal 1 can correctly determine that the object 2 is not a living body.

[0151] FIG. 17A is a diagram for explaining the characteristics of the infrared image of the eye of the printed matter (2) 5002. FIG. 17A includes the infrared image of the eye of the printed matter (2) 5002, a schematic diagram showing the state of the eye of the printed matter (2) 5002, and a histogram of the pixel values of the inscribed rectangular area of the iris area of the infrared image of the eye of the printed matter (2) 5002. The printed matter (2) 5002 is a fake image of a person, and it is different from the printed matter (1) 5001 in that there is one corneal reflection image in the eye of the person in the fake image.

[0152] In the infrared image of the eye of the printed matter (2) 5002, there is one corneal reflection. Furthermore, in the histogram of the infrared image of the eye of the printed matter (2) 5002, there are pixels having a relatively bright pixel value indicating one corneal reflection, and pixels having a relatively dark pixel value indicating pupil absorption.

[0153] FIG. 17B is a diagram for explaining the characteristics of the visible light image of the eye of the printed matter (2) 5002. FIG. 17B includes the visible light image of the eye of the printed matter (2) 5002, a schematic diagram showing the state of the eye of the printed matter (2) 5002, and a histogram of the pixel values of the inscribed rectangular area of the iris area of the visible light image of the eye of the printed matter (2) 5002.

[0154] In the visible light image of the eye of the printed matter (2) 5002, there is one corneal reflection. Furthermore, in the histogram of the visible light image of the eye of the printed matter (2) 5002, there are pixels having a relatively bright pixel value indicating one corneal reflection, and pixels having a relatively dark pixel value indicating pupil absorption.

[0155] When the object 2 is the printed matter (2) 5002, there is only one corneal reflection in the infrared image of the eye, and there is only one corneal reflection in the visible light image of the eye. Therefore, the biological determination condition is not satisfied. Thus, the information processing terminal 1 can correctly determine that the object 2 is not a living body even if it is a fake image of a person with a corneal reflection image in the eye like the printed matter (2) 5002.

[0156] <Specific operation> FIG. 18 is a flowchart showing a processing flow executed by the biological determination unit 105. The biological determination unit 105 starts processing from step 1800 and proceeds to step 1801, where the pupil absorption determination unit 103 determines whether there is pupil absorption in the infrared image of the eye of the object 2.

[0157] If there is no pupil absorption in the infrared image of the eye of the object 2, it is highly likely that the object 2 is not a living body. Therefore, in this case, the biological determination unit 105 determines "No" in step 1801, proceeds to step 1802, determines that the object 2 is not a living body, proceeds to step 1895, and temporarily ends this processing flow.

[0158] If there is pupil absorption in the infrared image of the eye of the object 2, in this determination, it is highly likely that the object 2 is a living body (the possibility of not being a living body is low). Therefore, in this case, the biological determination unit 105 determines "Yes" in step 1801, proceeds to step 1803, and the pupil absorption determination unit 103 determines whether it is determined that there is pupil absorption in the visible light image of the eye of the object 2.

[0159] If there is no pupil absorption in the visible light image of the eye of the object 2, it is highly likely that the object 2 is not a living body. Therefore, in this case, the biological determination unit 105 determines "No" in step 1803, proceeds to step 1802, determines that the object 2 is not a living body, proceeds to step 1895, and temporarily ends this processing flow.

[0160] When there is pupil absorption in the visible light image of the eyes of the subject 2, in this determination, the possibility that the subject 2 is a living body is high (the possibility that it is not a living body is low). Therefore, in this case, the living body determination unit 105 determines "Yes" in step 1803 and proceeds to step 1804. The light source control unit 115 drives two infrared light sources 10 and proceeds to step 1805.

[0161] When the living body determination unit 105 proceeds to step 1805, the corneal reflection determination unit 104 determines whether there are two corneal reflections in the infrared image of the eyes of the subject 2.

[0162] When there are not two corneal reflections in the infrared image of the eyes of the subject 2, the possibility that the subject 2 is not a living body is high. Therefore, in this case, the living body determination unit 105 determines "No" in step 1805, proceeds to step 1802, determines that the subject 2 is not a living body, proceeds to step 1895, and once ends this processing flow.

[0163] When there are two corneal reflections in the infrared image of the eyes of the subject 2, in this determination, the possibility that the subject 2 is a living body is high (the possibility that it is not a living body is low). Therefore, in this case, the living body determination unit 105 determines "Yes" in step 1805 and proceeds to step 1806. The visible light source 16 is driven by two and proceeds to step 1807.

[0164] When the living body determination unit 105 proceeds to step 1807, the corneal reflection determination unit 104 determines whether there are two corneal reflections in the visible light image of the eyes of the subject 2.

[0165] When there is no corneal reflection in the visible light image of the eyes of the subject 2, the possibility that the subject 2 is not a living body is high. Therefore, in this case, the living body determination unit 105 determines "No" in step 1807, proceeds to step 1802, determines that the subject 2 is not a living body, proceeds to step 1895, and once ends this processing flow.

[0166] When there are two corneal reflections in the visible light image of the eyes of the subject 2, in this determination, the possibility that the subject 2 is a living body is high (the possibility that it is not a living body is low). Therefore, in this case, the living body determination unit 105 determines "Yes" in step 1807 and proceeds to step 1808.

[0167] When the living body determination unit 105 proceeds to step 1808, there is pupil absorption in the infrared image of the eyes of the subject 2 ( "Yes" in step 1801), and there is pupil absorption in the visible light image of the eyes of the subject 2 ( "Yes" in step 1803), and there are two corneal reflections in the infrared image of the eyes of the subject 2 ( "Yes" in step 1805), and there are two corneal reflections in the visible light image of the eyes of the subject 2 ( "Yes" in step 1807). Since the living body determination conditions are satisfied, the possibility that the subject 2 is a living body becomes considerably high. Therefore, the living body determination unit 105 determines in step 1808 that the subject 2 is a living body, proceeds to step 1895, and once terminates this processing flow.

[0168] <Effect> As described above, the information processing terminal 1 according to the fourth embodiment of the present invention can accurately determine whether the subject 2 is a living body. When the subject 2 is a fake image (printed matter (1) 5001) of a person without a corneal reflection image in the eyes as shown in FIG. 16C, the information processing terminal 1 according to the fourth embodiment can accurately determine that the subject 2 is not a living body. Even when the subject 2 is a fake image (printed matter (2) 5002) of a person with a corneal reflection image in the eyes as shown in FIG. 17C, the information processing terminal 1 according to the fourth embodiment can accurately determine that the subject 2 is not a living body.

[0169] <<Fifth Embodiment>> The information processing terminal 1 according to the fifth embodiment of the present invention will be described. The information processing terminal 1 according to the fifth embodiment has differences from the information processing terminal 1 according to the fourth embodiment only in the following points. · The information processing terminal 1 according to the fifth embodiment changes the number of light emissions (lighting numbers) of the infrared light source 10, and determines whether there are corneal reflections in the infrared image of the eye of the subject 2 in a number (the same number as the number of light emissions) corresponding to the number of light emissions (lighting numbers). The information processing terminal 1 according to the fifth embodiment changes the number of light emissions (lighting numbers) of the visible light source 16, and determines whether there are corneal reflections in the visible light image of the eye of the subject 2 in a number (the same number as the number of light emissions) corresponding to the number of light emissions (lighting numbers). The information processing terminal 1 according to the fifth embodiment adopts a biometric determination condition different from the biometric determination condition of the fourth embodiment as the biometric determination condition.

[0170] Hereinafter, the description will focus on this difference.

[0171] <Overview> When the subject 2 is a living body, two corneal reflections exist in the infrared image of the eye of the subject 2 captured in a state where the eye of the subject 2 is illuminated by two infrared light sources 10. When the subject 2 is a living body, one corneal reflection exists in the infrared image of the eye of the subject 2 captured in a state where the eye of the subject 2 is illuminated by one infrared light source 10.

[0172] When the subject 2 is a living body, two corneal reflections exist in the visible light image of the eye of the subject 2 captured in a state where the eye of the subject 2 is illuminated by two visible light sources 16. When the subject 2 is a living body, one corneal reflection exists in the visible light image of the eye of the subject 2 captured in a state where the eye of the subject 2 is illuminated by one visible light source 16.

[0173] That is, when the subject 2 is a living body, if corneal reflections corresponding to the number of light emissions (lighting numbers) of the infrared light source 10 exist in the infrared image of the eye of the subject 2, and corneal reflections corresponding to the number of light emissions (lighting numbers) of the visible light source 16 exist in the visible light image of the eye of the subject 2, the possibility that the subject 2 is a living body becomes quite high.

[0174] Therefore, the information processing terminal 1 according to the fifth embodiment adopts, as a biological determination condition, that there is pupil absorption in the infrared image of the eye of the subject 2, and there is pupil absorption in the visible light image of the eye of the subject 2, and there are two corneal reflections in the infrared image of the eye of the subject 2 captured in a state where the eye of the subject 2 is illuminated by two infrared light sources 10, and there is one corneal reflection in the infrared image of the eye of the subject 2 captured in a state where the eye of the subject 2 is illuminated by one infrared light source 10, and there are two corneal reflections in the visible light image of the eye of the subject 2 captured in a state where the eye of the subject 2 is illuminated by two visible light sources 16, and there is one corneal reflection in the visible light image of the eye of the subject 2 captured in a state where the eye of the subject 2 is illuminated by one visible light source 16.

[0175] <Specific operation> FIG. 19 is a flowchart showing a processing flow executed by the biological determination unit 105. The biological determination unit 105 starts processing from step 1900 and proceeds to step 1901, where the pupil absorption determination unit 103 determines whether there is pupil absorption in the infrared image of the eye of the subject 2.

[0176] If there is no pupil absorption in the infrared image of the eye of the subject 2, it is highly likely that the subject 2 is not a living body. Therefore, in this case, the biological determination unit 105 determines "No" in step 1901, proceeds to step 1902, determines that the subject 2 is not a living body, proceeds to step 1995, and temporarily ends this processing flow.

[0177] If there is pupil absorption in the infrared image of the eye of the subject 2, in this determination, it is highly likely that the subject 2 is a living body (the possibility of not being a living body is low). Therefore, in this case, the biological determination unit 105 determines "Yes" in step 1901, proceeds to step 1903, and the pupil absorption determination unit 103 determines whether it has been determined that there is pupil absorption in the visible light image of the eye of the subject 2.

[0178] When there is no pupil absorption in the visible light image of the eyes of the subject 2, it is highly likely that the subject 2 is not a living body. Therefore, in this case, the living body determination unit 105 determines "No" in step 1903, proceeds to step 1902, determines that the subject 2 is not a living body, proceeds to step 1995, and temporarily ends this processing flow.

[0179] When there is pupil absorption in the visible light image of the eyes of the subject 2, in this determination, it is highly likely that the subject 2 is a living body (the possibility of not being a living body is low). Therefore, in this case, the living body determination unit 105 determines "Yes" in step 1903, proceeds to step 1904, drives two infrared light sources 10 by the light source control unit 115, and proceeds to step 1905.

[0180] When the living body determination unit 105 proceeds to step 1905, the corneal reflection determination unit 104 determines whether there are two corneal reflections in the infrared image of the eyes of the subject 2.

[0181] When there are not two corneal reflections in the infrared image of the eyes of the subject 2, it is highly likely that the subject 2 is not a living body. Therefore, in this case, the living body determination unit 105 determines "No" in step 1905, proceeds to step 1902, determines that the subject 2 is not a living body, proceeds to step 1995, and temporarily ends this processing flow.

[0182] When there are two corneal reflections in the infrared image of the eyes of the subject 2, in this determination, it is highly likely that the subject 2 is a living body (the possibility of not being a living body is low). Therefore, in this case, the living body determination unit 105 determines "Yes" in step 1905, proceeds to step 1906, drives one infrared light source 10, and proceeds to step 1907.

[0183] When the living body determination unit 105 proceeds to step 1907, it determines whether there is one corneal reflection in the infrared image of the eyes of the subject 2.

[0184] When there is no corneal reflex in the infrared image of the eye of the subject 2, it is highly likely that the subject 2 is not a living body. Therefore, in this case, the living body determination unit 105 determines "No" in step 1907, proceeds to step 1902, determines that the subject 2 is not a living body, proceeds to step 1995, and once ends this processing flow.

[0185] When there is one corneal reflex in the infrared image of the eye of the subject 2, in this determination, it is highly likely that the subject 2 is a living body (the possibility of not being a living body is low). Therefore, in this case, the living body determination unit 105 determines "Yes" in step 1907, proceeds to step 1908, drives two visible light sources 16, and proceeds to step 1909.

[0186] When proceeding to step 1909, the living body determination unit 105 determines whether there are two corneal reflexes in the visible light image of the eye of the subject 2.

[0187] When there are no two corneal reflexes in the visible light image of the eye of the subject 2, it is highly likely that the subject 2 is not a living body. Therefore, in this case, the living body determination unit 105 determines "No" in step 1909, proceeds to step 1902, determines that the subject 2 is not a living body, proceeds to step 1995, and once ends this processing flow.

[0188] When there are two corneal reflexes in the visible light image of the eye of the subject 2, in this determination, it is highly likely that the subject 2 is a living body (the possibility of not being a living body is low). Therefore, in this case, the living body determination unit 105 determines "Yes" in step 1909, proceeds to step 1910, drives one visible light source 16, and proceeds to step 1911.

[0189] When proceeding to step 1911, the living body determination unit 105 determines whether there is one corneal reflex in the visible light image of the eye of the subject 2.

[0190] When there is no corneal reflex in the visible light image of the eye of the subject 2, it is highly likely that the subject 2 is not a living body. Therefore, in this case, the living body determination unit 105 determines "No" in step 1911, proceeds to step 1902, determines that the subject 2 is not a living body, proceeds to step 1995, and once ends this processing flow.

[0191] When there is one corneal reflex in the visible light image of the eye of the subject 2, in this determination, it is highly likely that the subject 2 is a living body (the possibility of not being a living body is low). Therefore, in this case, the living body determination unit 105 determines "Yes" in step 1911 and proceeds to step 1912.

[0192] When the living body determination unit 105 proceeds to step 1912, there is pupil absorption in the infrared image of the eye of the subject 2 ( "Yes" in step 1901), and there is pupil absorption in the visible light image of the eye of the subject 2 ( "Yes" in step 1903), and there are two corneal reflexes in the infrared image of the eye of the subject 2 captured in a state where the eye of the subject 2 is illuminated by two infrared light sources 10 ( "Yes" in step 1905), there is one corneal reflex in the infrared image of the eye of the subject 2 captured in a state where the eye of the subject 2 is illuminated by one infrared light source 10 ( "Yes" in step 1907), and there are two corneal reflexes in the visible light image of the eye of the subject 2 captured in a state where the eye of the subject 2 is illuminated by two visible light sources 16 ( "Yes" in step 1909), and there is one corneal reflex in the visible light image of the eye of the subject 2 captured in a state where the eye of the subject 2 is illuminated by one visible light source 16 ( "Yes" in step 1911). Since the living body determination conditions are satisfied, the possibility that the subject 2 is a living body becomes quite high. Therefore, the living body determination unit 105 determines in step 1912 that the subject 2 is a living body, proceeds to step 1995, and once ends this processing flow.

[0193] <Effect> As described above, the information processing terminal 1 according to the fifth embodiment of the present invention can accurately determine whether the subject 2 is a living body. When the subject 2 is a fake image (printed matter (1) 5001) of a person without a corneal reflection image in the eyes as shown in FIG. 16C, the information processing terminal 1 according to the fifth embodiment can accurately determine that the subject 2 is not a living body. Even when the subject 2 is a fake image (printed matter (2) 5002) of a person with a corneal reflection image in the eyes as shown in FIG. 17C, the information processing terminal 1 according to the fifth embodiment can accurately determine that the subject 2 is not a living body.

[0194] <<Sixth Embodiment>> The information processing terminal 1 according to the sixth embodiment of the present invention will be described. The information processing terminal 1 according to the sixth embodiment has differences from the information processing terminal 1 according to the fourth embodiment only in the following points. · The information processing terminal 1 according to the sixth embodiment changes the number of light emissions (number of lit lights) of the visible light source 16 and determines whether the pupil diameter changes (the size of the pupil changes) according to the number of light emissions (number of lit lights). The information processing terminal 1 according to the sixth embodiment adopts a living body determination condition different from the living body determination condition of the fourth embodiment as the living body determination condition.

[0195] Hereinafter, the description will be centered on this difference.

[0196] <Configuration> FIG. 20 is a diagram showing a functional block of the controller 11 of the information processing terminal 1 according to the sixth embodiment. As shown in FIG. 20, the controller 11 of the information processing terminal 1 according to the sixth embodiment includes a pupil diameter measurement unit 106. The pupil diameter measurement unit 106 takes the image of the eye output by the eye detection unit 102 as an input and measures the pupil diameter from the image of the eye. The pupil diameter measurement unit 106 outputs the measured pupil diameter to the living body determination unit 105. The living body determination unit 105 determines whether the pupil diameter has expanded (the size of the pupil has changed) based on the input pupil diameter (this determination may be referred to as "pupil size determination"), and performs living body determination using the determination result.

[0197] <Overview> When the subject 2 is a living body, two corneal reflections exist in the infrared image of the eye of the subject 2 captured in a state where the eyes of the subject 2 are illuminated by two infrared light sources 10. When the subject 2 is a living body, one corneal reflection exists in the infrared image of the eye of the subject 2 captured in a state where the eyes of the subject 2 are illuminated by one infrared light source 10.

[0198] When the subject 2 is a living body, when the light is scarce (weak), the pupil expands and becomes larger, and when the light is abundant (strong), the pupil contracts and becomes smaller. Therefore, the pupil diameter in a state where the eyes of the subject 2 are illuminated by one visible light source 16 is larger than the pupil diameter in a state where the eyes of the subject 2 are illuminated by two visible light sources 16.

[0199] That is, when the subject 2 is a living body, a corneal reflection corresponding to the number of light-emitting (lit) infrared light sources 10 exists in the infrared image of the eye of the subject 2. If the pupil diameter in a state where the eyes of the subject 2 are illuminated by one visible light source 16 is larger than the pupil diameter in a state where the eyes of the subject 2 are illuminated by two visible light sources 16, the possibility that the subject 2 is a living body becomes quite high.

[0200] Therefore, the information processing terminal 1 according to the sixth embodiment employs, as a biological determination condition, that there is pupil absorption in the infrared image of the eye of the subject 2, there is pupil absorption in the visible light image of the eye of the subject 2, there are two corneal reflections in the infrared image of the eye of the subject 2 captured in a state where the eyes of the subject 2 are illuminated by two infrared light sources 10, there is one corneal reflection in the infrared image of the eye of the subject 2 captured in a state where the eyes of the subject 2 are illuminated by one infrared light source 10, and the pupil diameter in a state where the eyes of the subject 2 are illuminated by one visible light source 16 is larger than the pupil diameter in a state where the eyes of the subject 2 are illuminated by two visible light sources 16 (the pupil diameter has changed).

[0201] <Specific operation> FIG. 21 is a flowchart showing the processing flow executed by the biological determination unit 105. The biological determination unit 105 starts processing from step 2100 and proceeds to step 2101, where the pupil absorption determination unit 103 determines whether there is pupil absorption in the infrared image of the eye of the subject 2.

[0202] If there is no pupil absorption in the infrared image of the eye of the subject 2, it is highly likely that the subject 2 is not a living body. Therefore, in this case, the biological determination unit 105 determines "No" in step 2101, proceeds to step 2102, determines that the subject 2 is not a living body, proceeds to step 2195, and temporarily ends this processing flow.

[0203] If there is pupil absorption in the infrared image of the eye of the subject 2, in this determination, it is highly likely that the subject 2 is a living body (the possibility that the subject 2 is not a living body is low). Therefore, in this case, the biological determination unit 105 determines "Yes" in step 2101, proceeds to step 2103, and the pupil absorption determination unit 103 determines whether it is determined that there is pupil absorption in the visible light image of the eye of the subject 2.

[0204] If there is no pupil absorption in the visible light image of the eye of the subject 2, it is highly likely that the subject 2 is not a living body. Therefore, in this case, the biological determination unit 105 determines "No" in step 2103, proceeds to step 2102, determines that the subject 2 is not a living body, proceeds to step 2195, and temporarily ends this processing flow.

[0205] If there is pupil absorption in the visible light image of the eye of the subject 2, in this determination, it is highly likely that the subject 2 is a living body (the possibility that it is not a living body is low). Therefore, in this case, the biological determination unit 105 determines "Yes" in step 2103, proceeds to step 2104, the light source control unit 115 drives two infrared light sources 10, and proceeds to step 2105.

[0206] When the biological determination unit 105 proceeds to step 2105, the corneal reflection determination unit 104 determines whether there are two corneal reflections in the infrared image of the eye of the subject 2.

[0207] When there are no two corneal reflections in the infrared image of the eyes of the subject 2, there is a high possibility that the subject 2 is not a living body. Therefore, in this case, the living body determination unit 105 determines "No" in step 2105, proceeds to step 2102, determines that the subject 2 is not a living body, proceeds to step 2195, and once ends this processing flow.

[0208] When there are two corneal reflections in the infrared image of the eyes of the subject 2, in this determination, there is a high possibility that the subject 2 is a living body (the possibility of not being a living body is low). Therefore, in this case, the living body determination unit 105 determines "Yes" in step 2105, proceeds to step 2106, drives one infrared light source 10, and proceeds to step 2107.

[0209] When proceeding to step 2107, the living body determination unit 105 determines whether there is one corneal reflection in the infrared image of the eyes of the subject 2.

[0210] When there is no one corneal reflection in the infrared image of the eyes of the subject 2, there is a high possibility that the subject 2 is not a living body. Therefore, in this case, the living body determination unit 105 determines "No" in step 2107, proceeds to step 2102, determines that the subject 2 is not a living body, proceeds to step 2195, and once ends this processing flow.

[0211] When there is one corneal reflection in the infrared image of the eyes of the subject 2, in this determination, there is a high possibility that the subject 2 is a living body (the possibility of not being a living body is low). Therefore, in this case, the living body determination unit 105 determines "Yes" in step 2107, and after sequentially executing the processes of steps 2108 to 2111 described below, proceeds to step 2112.

[0212] Step 2108: The living body determination unit 105 drives two visible light sources 16 by the light source control unit 115.

[0213] Step 2109: The living body determination unit 105 measures the pupil diameter by the pupil diameter measurement unit 106.

[0214] Step 2110: The biological determination unit 105 drives one visible light source 16 by the light source control unit 115.

[0215] Step 2111: The biological determination unit 105 measures the pupil diameter by the pupil diameter measurement unit 106.

[0216] When the biological determination unit 105 proceeds to step 2112, it determines whether the pupil diameter measured with the eye of the subject 2 illuminated by one visible light source 16 is larger than the pupil diameter measured with the eye of the subject 2 illuminated by two visible light sources 16, thereby determining whether the pupil diameter has expanded.

[0217] If the pupil diameter has not expanded, it is highly likely that the subject 2 is not a living body. Therefore, in this case, the biological determination unit 105 determines "No" in step 2112, proceeds to step 2102, determines that the subject 2 is not a living body, and then proceeds to step 2195 to temporarily end this processing flow.

[0218] If the pupil diameter has expanded, the biological determination unit 105 determines "Yes" in step 2112 and proceeds to step 2113.

[0219] When the biometric determination unit 105 proceeds to step 2112, there is pupil absorption in the infrared image of the eye of the subject 2 (Yes in step 2101), and there is pupil absorption in the visible light image of the eye of the subject 2 (Yes in step 2103), and there are two corneal reflections in the infrared image of the eye of the subject 2 captured while illuminating the eye of the subject 2 with two infrared light sources 10 (Yes in step 2105), there is one corneal reflection in the infrared image of the eye of the subject 2 captured while illuminating the eye of the subject 2 with one infrared light source 10 (Yes in step 2107), and the pupil diameter in the state where the eye of the subject 2 is illuminated by one visible light source 16 is larger than the pupil diameter in the state where the eye of the subject 2 is illuminated by two visible light sources 16 (Yes in step 2112). Since the biometric determination condition is satisfied, the possibility that the subject 2 is a living body becomes quite high. Therefore, the biometric determination unit 105 determines in step 2113 that the subject 2 is a living body and proceeds to step 2195 to temporarily end this processing flow.

[0220] <Effect> As described above, the information processing terminal 1 according to the sixth embodiment of the present invention can accurately determine whether the subject 2 is a living body. When the subject 2 is a fake image (printed matter (1) 5001) of a person without a corneal reflection image in the eye as shown in FIG. 16C, the information processing terminal 1 according to the sixth embodiment can accurately determine that the subject 2 is not a living body. Even when the subject 2 is a fake image (printed matter (2) 5002) of a person with a corneal reflection image in the eye as shown in FIG. 17C, the information processing terminal 1 according to the sixth embodiment can accurately determine that the subject 2 is not a living body.

[0221] <<Modification Example>> The present invention is not limited to the above-described embodiments, and various modification examples can be adopted within the scope of the present invention. Furthermore, the above-described embodiments can be combined with each other as long as they do not depart from the scope of the present invention.

[0222] In the above-described second embodiment and the above-described third embodiment, the number of infrared light sources 10 is two, but the number of infrared light sources 10 may be three or more. In the above-described fourth embodiment to the above-described sixth embodiment, the number of infrared light sources 10 is two, but the number of infrared light sources 10 may be three or more. In the above-described fourth embodiment to the above-described sixth embodiment, the number of visible light sources 16 is two, but the number of visible light sources 16 may be three or more.

[0223] In the above-described sixth embodiment, the process of step 2110 in FIG. 21 may be replaced with a process of turning off the driving of the visible light source 16. In the above-described sixth embodiment, the processes of step 2106 and step 2107 may be omitted. In the above-described sixth embodiment, the processes of step 2104 and step 2105 may be omitted.

[0224] In the above-described sixth embodiment, the process of step 2108 in FIG. 21 may be changed to a process of driving one visible light source 16, the process of step 2110 may be changed to a process of driving two visible light sources 16, and the process of step 2112 may be changed to a process of determining whether or not the pupil diameter has decreased. In this case, if it is determined at step 2112 that the pupil diameter has decreased, the process proceeds to step 2113, and it is determined that the subject 2 is a living body. If it is determined at step 2112 that the pupil diameter has not decreased, the process proceeds to step 2102, and it is determined that the subject 2 is a living body.

[0225] The features of each of the above-described first embodiment to the above-described fifth embodiment and the pupil size determination of the sixth embodiment may be combined. In this case, in each modification, a biological determination condition including each biological determination condition of the above-described first embodiment to the above-described fifth embodiment and that the pupil diameter has expanded (the size of the pupil has changed) is adopted.

[0226] In the above-described sixth embodiment, the visible light source 16 may be provided with only one, and it may be determined whether or not the size of the pupil changes based on whether or not the pupil diameter changes when one visible light source 16 is turned on from off or off from on.

[0227] The modified example of the third embodiment can also adopt the following configuration.

[0228] A plurality of infrared light sources that generate infrared light and illuminate the eyes of a subject that is the object of biological determination, An infrared camera that images the subject, A visible light camera that images the subject, A controller that controls the infrared camera and the visible light camera, A biological determination device comprising: The controller: In the first corneal reflex determination, determine the number of corneal reflexes when there is a corneal reflex in the eye of the subject, After lighting the infrared light sources a predetermined number of times (in the third embodiment, it was 2, but it may be other than 2), image the subject with the infrared camera, perform the first corneal reflex determination, then change the number of lit infrared light sources from the previous number of lit sources, image the subject with the infrared camera, and perform the first corneal reflex determination. Do this a predetermined number of times (in the third embodiment, it was 1, but it may be 2 or more), As the biological determination conditions, In the determination result of the first pupil absorption determination, there is pupil absorption in the eye of the subject, and in the first corneal reflex determination, there is a corneal reflex corresponding to the number of lit infrared light sources, and In the determination results of the second pupil absorption determination and the second corneal reflex determination, there is pupil absorption in the eye of the subject, and there is no corneal reflex, Determine whether it holds, Is configured as A biological determination device.

[0229] The modified example of the fifth embodiment can also adopt the following configuration.

[0230] A plurality of infrared light sources that generate infrared light and illuminate the eyes of a subject that is the object of biological determination, A plurality of visible light sources that generate visible light and illuminate the eyes of the subject, An infrared camera that images the subject, A visible light camera that images the subject, A controller that controls the infrared camera and the visible light camera, A living body determination device comprising: The controller, In the first corneal reflex determination, determines the number of corneal reflexes when there is a corneal reflex in the eye of the subject, After lighting the infrared light source a predetermined number of times (although it was two in the fifth embodiment, it may be other than two) and imaging the subject with the infrared camera and performing the first corneal reflex determination, changes the number of lighting of the infrared light source from the previous number of lighting and images the subject with the infrared camera and performs the first corneal reflex determination, and performs this a predetermined number of times (although it was one in the fifth embodiment, it may be two or more times), In the second corneal reflex determination, determines the number of corneal reflexes when there is a corneal reflex in the eye of the subject, After lighting the visible light source a predetermined number of times (although it was two in the fifth embodiment, it may be other than two) and imaging the subject with the visible light camera and performing the second corneal reflex determination, changes the number of lighting of the visible light source from the previous number of lighting and images the subject with the visible light camera and performs the second corneal reflex determination, and performs this a predetermined number of times (although it was one in the fifth embodiment, it may be two or more times), As the living body determination condition, In the determination results of the first pupil absorption determination and the first corneal reflex determination, there is pupil absorption in the eye of the subject, and there is a corneal reflex corresponding to the number of lighting of the infrared light source, and In the determination results of the second pupil absorption determination and the second corneal reflex determination, there is pupil absorption in the eye of the subject, and there is a corneal reflex corresponding to the number of lighting of the visible light source, Determines whether or not it holds, Is configured as Living body determination device.

Explanation of Signs

[0231] 1... Information processing terminal, 10... Infrared light source, 11... Controller, 12... Visible light camera, 13... Infrared camera, 14... Display, 102... Eye detection unit, 103... Pupil absorption determination unit, 104... Corneal reflection determination unit, 105... Living body determination unit, 111... CPU, 112... ROM, 113... RAM, 114... Camera processing unit, 115... Light source control unit, 116... Display control unit

Claims

1. An infrared light source that generates infrared light and illuminates the eyes of a subject to be biometrically determined, An infrared camera that images the subject, A visible light camera that images the subject, A controller that controls the infrared camera and the visible light camera, A biometric determination device comprising: The controller: Detects the eyes of the subject from the infrared image of the subject captured by the infrared camera, and performs a first pupil absorption determination to determine whether there is pupil absorption in the eyes of the subject, and a first corneal reflection determination to determine whether there is corneal reflection in the eyes of the subject; Detects the eyes of the subject from the visible light image of the subject captured by the visible light camera, and performs a second pupil absorption determination to determine whether there is pupil absorption in the eyes of the subject, and a second corneal reflection determination to determine whether there is corneal reflection in the eyes of the subject; Based on the determination results of the first pupil absorption determination and the first corneal reflection determination, and the second pupil absorption determination and the second corneal reflection determination, determines whether a biometric determination condition, which is a condition indicating that the subject is a living body, is satisfied; When the biometric determination condition is satisfied, determines that the subject is a living body; When the biometric determination condition is not satisfied, determines that the subject is not a living body. A biometric determination device configured as described above. Biometric determination device.

2. In the biometric determination device according to Claim 1, The controller: As the biometric determination condition, Determines whether it is satisfied that in the determination results of the first pupil absorption determination and the first corneal reflection determination, there is pupil absorption in the eyes of the subject, and there is corneal reflection, and in the determination results of the second pupil absorption determination and the second corneal reflection determination, there is pupil absorption in the eyes of the subject, and there is no corneal reflection; A biometric determination device configured as described above. Biometric determination device. Biometric determination device.

3. In the biometric determination device according to Claim 1, Comprises a plurality of the infrared light sources, The controller: In the first corneal reflection determination, determines the number of corneal reflections when there is corneal reflection in the eyes of the subject; As the biometric determination condition, Determines whether it is satisfied that in the determination results of the first pupil absorption determination and the first corneal reflection determination, there is pupil absorption in the eyes of the subject, and there are the same number of corneal reflections as the number of the infrared light sources, and In the determination results of the second pupil absorption determination and the second corneal reflection determination, there is pupil absorption in the eyes of the subject, and there is no corneal reflection; A biometric determination device configured as described above. configured as follows, biometric authentication device.

4. In the biometric authentication device according to claim 1, comprising a plurality of the infrared light sources, the controller, in the first corneal reflection determination, determines the number of corneal reflections when there is a corneal reflection in the eye of the subject, after lighting the infrared light source a predetermined number of times and imaging the subject with the infrared camera to perform the first corneal reflection determination, changes the number of lightings of the infrared light source from the previous number of lightings, images the subject with the infrared camera, and performs the first corneal reflection determination, and repeats this a predetermined number of times, as the biometric authentication condition, in the determination results of the first pupil absorption determination and the first corneal reflection determination, there is pupil absorption in the eye of the subject, and there is a corneal reflection corresponding to the number of lightings of the infrared light source, and in the determination results of the second pupil absorption determination and the second corneal reflection determination, there is pupil absorption in the eye of the subject, and there is no corneal reflection, determines whether or not it is satisfied, configured as follows, biometric authentication device.

5. In the biometric authentication device according to claim 1, a plurality of the infrared light sources, and a plurality of visible light sources that generate visible light and illuminate the eyes of the subject, are provided, the controller, in the first corneal reflection determination, determines the number of corneal reflections when there is a corneal reflection in the eye of the subject, in the second corneal reflection determination, determines the number of corneal reflections when there is a corneal reflection in the eye of the subject, as the biometric authentication condition, in the determination results of the first pupil absorption determination and the first corneal reflection determination, there is pupil absorption in the eye of the subject, and there are the same number of corneal reflections as the number of the infrared light sources, and in the determination results of the second pupil absorption determination and the second corneal reflection determination, there is pupil absorption in the eye of the subject, and there are the same number of corneal reflections as the number of the visible light sources, determines whether or not it is satisfied, configured as follows, biometric authentication device.

6. In the biometric authentication device according to claim 1, a plurality of the infrared light sources, and a plurality of visible light sources that generate visible light and illuminate the eyes of the subject, are provided, the controller, in the first corneal reflection determination, determines the number of corneal reflections when there is a corneal reflection in the eye of the subject, The infrared light source is turned on a predetermined number of times, and the subject is imaged by the infrared camera. After performing the first corneal reflex determination, the number of times the infrared light source is turned on is changed from the previous number of times, and the subject is imaged by the infrared camera, and the first corneal reflex determination is performed a predetermined number of times. In the second corneal reflex determination, the number of corneal reflexes when there is a corneal reflex in the eye of the subject is determined. The visible light source is turned on a predetermined number of times, and the subject is imaged by the visible light camera. After performing the second corneal reflex determination, the number of times the visible light source is turned on is changed from the previous number of times, and the subject is imaged by the visible light camera, and the second corneal reflex determination is performed a predetermined number of times. As the biological determination conditions, In the determination results of the first pupil absorption determination and the first corneal reflex determination, there is pupil absorption in the eye of the subject, and there is a corneal reflex corresponding to the number of times the infrared light source is turned on, and In the determination results of the second pupil absorption determination and the second corneal reflex determination, there is pupil absorption in the eye of the subject, and there is a corneal reflex corresponding to the number of times the visible light source is turned on. It is determined whether or not it is satisfied. configured as biological determination device.

7. In the biological determination device according to claim 1, A visible light source that generates visible light and illuminates the eye of the subject is provided. The controller changes the number of times the visible light source is turned on, and performs a pupil size determination to determine whether the size of the pupil has changed when the number of times the visible light source is turned on is changed. Based on the determination results of the first pupil absorption determination, the first corneal reflex determination, the second pupil absorption determination, the second corneal reflex determination, and the pupil size determination, it is determined whether the biological determination conditions are satisfied. configured as biological determination device.

8. In the biological determination device according to claim 1, The optical axis of the infrared light source is different from the optical axis of the infrared camera, and the optical axis of the infrared light source is different from the optical axis of the visible light camera. biological determination device.

9. An infrared light source that generates infrared light and illuminates the eye of a subject that is a biological determination target, A visible light source that generates visible light and illuminates the eye of the subject, An infrared camera that images the subject, A visible light camera that images the subject, A controller that controls the infrared camera and the visible light camera, A biological determination device comprising: The controller Detect the eyes of the subject from the infrared image of the subject captured by the infrared camera, and perform a first pupil absorption determination to determine whether there is pupil absorption in the eyes of the subject, and a first corneal reflection determination to determine whether there is corneal reflection in the eyes of the subject. Detect the eyes of the subject from the visible light image of the subject captured by the visible light camera, perform a second pupil absorption determination to determine whether there is pupil absorption in the eyes of the subject, change the number of times the visible light source is turned on, and perform a pupil size determination to determine whether the size of the pupil has changed when the number of times the visible light source is turned on is changed. Based on the determination results of the first pupil absorption determination and the first corneal reflection determination, and the second pupil absorption determination and the pupil size determination, determine whether a living body determination condition, which is a condition indicating that the subject is a living body, is satisfied. When the living body determination condition is satisfied, determine that the subject is a living body. When the living body determination condition is not satisfied, determine that the subject is not a living body. configured as a living body determination device.

10. In the living body determination device according to claim 9, the controller as the living body determination condition in the determination results of the first pupil absorption determination and the first corneal reflection determination, there is pupil absorption in the eyes of the subject, and there is corneal reflection, and in the determination result of the second pupil absorption determination, there is pupil absorption in the eyes of the subject, and in the determination result of the pupil size determination, the size of the pupil has changed, to determine whether it is satisfied. configured as a living body determination device.

11. In the living body determination device according to claim 9, equipped with a plurality of the infrared light sources and a plurality of the visible light sources, the controller in the first corneal reflection determination, determine the number of corneal reflections when there is corneal reflection in the eyes of the subject, turn on the infrared light source by a predetermined number of times to image the subject with the infrared camera, perform the first corneal reflection determination, then change the number of times the infrared light source is turned on from the previous number of times and image the subject with the infrared camera, and perform the first corneal reflection determination a predetermined number of times. as the living body determination condition in the determination results of the first pupil absorption determination and the first corneal reflection determination, there is pupil absorption in the eyes of the subject, and there is corneal reflection corresponding to the number of times the infrared light source is turned on, and In the determination result of the second pupil absorption determination, there is pupil absorption in the eye of the subject, and in the determination result of the pupil size determination, the size of the pupil has changed, determine whether or not it holds, Biometric authentication device.

12. An infrared light source that generates infrared light and illuminates the eye of a subject that is the object of biometric authentication, An infrared camera that images the subject, A visible light camera that images the subject, A controller that controls the infrared camera and the visible light camera, A biometric authentication method using the following, by the controller, Detect the eye of the subject from the infrared image of the subject captured by the infrared camera, and perform a first pupil absorption determination to determine whether there is pupil absorption in the eye of the subject, and a first corneal reflection determination to determine whether there is a corneal reflection in the eye of the subject, Detect the eye of the subject from the visible light image of the subject captured by the visible light camera, and perform a second pupil absorption determination to determine whether there is pupil absorption in the eye of the subject, and a second corneal reflection determination to determine whether there is a corneal reflection in the eye of the subject, Based on the determination results of the first pupil absorption determination and the first corneal reflection determination, and the second pupil absorption determination and the second corneal reflection determination, determine whether or not a biometric authentication condition, which is a condition indicating that the subject is a living body, is satisfied, When the biometric authentication condition is satisfied, determine that the subject is a living body, When the biometric authentication condition is not satisfied, determine that the subject is not a living body, Biometric authentication method.

13. An infrared light source that generates infrared light and illuminates the eye of a subject that is the object of biometric authentication, A visible light source that generates visible light and illuminates the eye of a subject that is the object of biometric authentication, An infrared camera that images the subject, A visible light camera that images the subject, A controller that controls the infrared camera and the visible light camera, A biometric authentication method using the following, The controller, Detect the eye of the subject from the infrared image of the subject captured by the infrared camera, and perform a first pupil absorption determination to determine whether there is pupil absorption in the eye of the subject, and a first corneal reflection determination to determine whether there is a corneal reflection in the eye of the subject, Detect the eyes of the subject from the visible light image of the subject captured by the visible light camera, and perform a second pupil absorption determination to determine whether there is pupil absorption in the eyes of the subject, and change the number of times the visible light source is turned on, and determine whether the size of the pupil changes when the number of times the visible light source is turned on is changed, Based on the determination results of the first pupil absorption determination and the first corneal reflection determination, and the second pupil absorption determination and the pupil size determination, determine whether a living body determination condition, which is a condition indicating that the subject is a living body, is satisfied, When the living body determination condition is satisfied, determine that the subject is a living body, When the living body determination condition is not satisfied, determine that the subject is not a living body, Living body determination method.

Images