JP2006099614A - Living body discrimination apparatus and living body discrimination method - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To provide a living body discrimination apparatus capable of gaining desired control without performing erroneous operation or erroneous discrimination even when a photograph, picture, video display equipment, and the like exist within a photographing range without requesting special biological response or operation of a figure in an apparatus using feature information or state obtained by photographing living body information on the figure, for example. <P>SOLUTION: In a living body discrimination apparatus for discriminating whether an object for which living body information (for example, a face image) is acquired is a living body or a non-living body, an image including the living body information on the object is obtained, a living body area (a face area) is detected from the acquired image, image characteristic in the detected living area is compared with that in the living body area obtained when a normal object obtained in advance exists at a prescribed position, and the object is discriminated to be non-living body when it is judged that there is a difference equal to a prescribed value or more between the image characteristics. <P>COPYRIGHT: (C)2006,JPO&NCIPI

Description

  The present invention, for example, is a device that records biometric information itself, such as the iris, retina, veins, facial features, and hand status of a person, and records the biometric information itself. In a user interface device that detects movement or the like, a biometric authentication device that determines whether or not the person is the person based on the biometric information, the target for acquiring the biometric information is a living body (real person) or a non-living body ( The present invention relates to a biological discrimination device and a biological discrimination method for discriminating whether an object other than a person).

  Recently, for example, a device that records biometric information itself, such as the iris, retina, veins, facial features, hand status, etc. Various techniques for biometric information have been developed, such as a user interface device that detects a biometric information and a biometric authentication device that determines whether or not the user is the person based on the biometric information.

  Among them, in a biometric authentication device that uses feature information and state obtained by photographing a person's biometric information, a non-biological object is printed by a photograph or picture printed on the area to be photographed or a picture display device. If the image is taken, there is a problem that the desired operation cannot be performed by the above-described device, and the malfunction or the erroneous determination is caused.

Therefore, recently, as means for solving these problems, a special biological reaction operation is performed on a person holding biological information to be imaged, or a predetermined operation is requested for a person holding the biological information. (For example, refer to Patent Document 1, Patent Document 2, and Patent Document 3).
Japanese Patent Laid-Open No. 11-339048 Japanese Patent Laid-Open No. 2003-242491 Japanese Patent Laid-Open No. 2000-30080

However, in the conventional method described above, a special biological reaction operation is performed on the person holding the biological information to be imaged, or a predetermined operation is requested from the person holding the biological information. There are many cases where a burden is required for the user.
In addition, there is a problem that a precise measuring device that can measure a change in the pupil of the eye and distinguish the unevenness of the face is required, and a device added to a normal device becomes complicated and expensive.

  Therefore, the present invention, for example, in an apparatus that uses characteristic information and state obtained by photographing biological information of a person, displays pictures, pictures, and images without requiring a special biological reaction or operation to the person. An object of the present invention is to provide a living body discriminating apparatus and a living body discriminating method capable of performing desired control without malfunction or misjudgment even when a device or the like is present in the imaging range.

  The living body discriminating apparatus of the present invention is a living body discriminating apparatus that discriminates whether a target for acquiring biological information is a living body or a non-living body, and an image acquiring unit that acquires an image including the biological information of the target object. A living body detecting means for detecting a living body area from the image acquired by the image acquiring means, and a normal object obtained in advance for the image features of the living body area detected by the living body detecting means at a predetermined position. A living body discriminating means for discriminating that the object is non-living when it is determined that there is a difference of a predetermined value or more between the image features of the living body region obtained when Output control means for changing contents to be output to a device connected as an output destination based on the determination result of the determination means.

  Further, the living body discrimination device of the present invention is a living body discrimination device that discriminates whether a target for which biological information is acquired is a living body or a non-living body, and acquires an image including biological information of the target. Means, a living body detecting means for detecting a living body area from the image acquired by the image acquiring means, and a living body in an area other than the living body area detected by the living body detecting means in the image acquired by the image acquiring means. Based on the determination result of the living body determining means, and the living body determining means for determining whether the target object is non-living when detecting the presence of an object other than the living body, Output control means for changing contents to be output to a device connected as an output destination.

  Further, the living body discrimination method of the present invention is a living body discrimination method for discriminating whether an object for obtaining biological information is a living body or a non-living body, and acquires an image including the biological information of the object. A living body region is detected from the acquired image, and the image feature of the detected living body region is compared with the image feature of the living body region obtained when a normal object obtained in advance is present at a predetermined position; When it is determined that there is a difference greater than or equal to a predetermined value between the two image features, the object is determined to be non-living.

  Further, the living body discrimination method of the present invention is a living body discrimination method for discriminating whether an object for obtaining biological information is a living body or a non-living body, and acquires an image including the biological information of the object. A living body region is detected from the acquired image, and it is detected whether an object other than the living body is present in a region other than the detected living body region in the acquired image, and the presence of an object other than the living body is detected. When detected, the object is determined to be non-living.

  According to the present invention, for example, in an apparatus that uses characteristic information and a state obtained by photographing biological information of a person, a photograph, a picture, and a video are displayed without requiring a special biological reaction or operation from the person. It is possible to provide a living body determination apparatus and a living body determination method capable of performing desired control without malfunction or erroneous determination even when a device or the like is present in the imaging range.

Hereinafter, embodiments of the present invention will be described with reference to the drawings.
First, the first embodiment will be described.
In the following description, an image is input by a camera as a person to be imaged (object from which biometric information is acquired), a face area is detected, and the object to be imaged is a real person (living body) or an object other than a person An example of a biological discrimination device that changes a control signal to a connected device by determining whether it is (non-living), but the present invention is not limited to a human face, It can be applied to those that perform biometric discrimination by inputting images containing various biometric information. For example, biometric discrimination using iris information, retinal information, vein information of hands and fingers, fingerprint information, eye and mouth status, etc. Is also applicable.

  FIG. 1 schematically shows a configuration of a living body discrimination apparatus according to the first embodiment. The biometric discrimination device includes a camera 101 as an image acquisition unit that captures and inputs a face image (an image including at least a face) of a person (hereinafter also referred to as a shooting target or an object) M to be shot. 101 from a face image of a person obtained from 101, a face detection unit 102 as a biometric detection unit for detecting the position of a face part such as a face, eyes, and nose, and an imaging target based on an image feature of the face area detected by the face detection unit 102 The biometric discrimination section 103 as a biometric discrimination means for discriminating whether M is a real living body or a printed matter such as a photograph or a picture or a video display device, and the face image or face of the person M photographed by the camera 101 Depending on the detection result of the face area detected by the detection unit 102 and the determination result of the living body determination unit 103, the connected device, the operating output signal, and the program control means are changed. And an output control section 104 as an output control unit that.

  The camera 101 captures and inputs a face image of the person M. The camera 101 is mainly composed of, for example, a television camera using an image sensor such as a CCD sensor, and A / D converts the captured face image. After being digitized, it is sent to the face detection unit 102.

  First, the face detection unit 102 detects the face area of the person M from the image input by the camera 101. For example, in the input image from the camera 101, the position that gives the highest correlation value is determined as the face region by obtaining the correlation value while moving a template prepared in advance in the image. In addition, face detection means such as a face region extraction method using an eigenspace method or a subspace method may be used.

  Next, the face detection unit 102 detects the position of a face part such as eyes, nose, and mouth from the detected face area. As the detection method, for example, literature (Kazuhiro Fukui, Osamu Yamaguchi: “Face feature point extraction by combination of shape extraction and pattern matching”, IEICE Transactions (D), vol. J80-D-II, No. 8, pp 2170-2177 (1997)) may be used.

  In the present embodiment, an example in which the feature of the face area is used has been described. However, for example, in the case of determining whether the subject is an iris, a retina, or an eye image, whether it is a living body or a non-living body, The detection can be performed by zooming the camera based on the eye area detected by the processing.

  In addition to the above eye and nose detection, the detection of the mouth area is described, for example, in the literature (Mayumi Yuasa, Akiko Nakajima: “Digital Make System Based on High-Precision Facial Feature Point Detection” Proc. It can be easily realized by using the technology of pp219-224 (2004).

  As for fingerprints and vein patterns, for example, literature (Optoelectronic Industry and Technology Promotion Association (http://www.oitda.or.jp/index-j.html): 2003 Optical Technology Trend Survey “15- This can be realized by using the input means and technology described in “003-1 Optical Technology Trend Survey Report”, Chapter 5 “Human Interface” (2003)). In any case, it is possible to acquire information that can be handled as a two-dimensional array image and detect a region of the biometric feature from the acquired information.

  The biometric discrimination unit 103 obtains in advance image-like feature information (image features) of a position where a person stands on average or a face area obtained when a person stands in a predetermined position based on a guide or the like in advance. Find the distribution. The image feature may be, for example, the size of the face image such as the distance between both eyes, the area obtained by multiplying the distance between both eyes and the distance between the eyes, or the luminance distribution using the average luminance in the face region, the intermediate luminance value, or the like. It is also possible to use a value indicating contrast, such as the sum of the differential values of the image in the face region or the ratio of the high frequency component multiplied by FFT.

  The biometric discrimination unit 103 obtains the above-mentioned image feature again from the biometric feature when input by the camera 101, compares it with the distribution of the image feature obtained for the normal person obtained in advance, and obtains it for the normal person. When a value other than a predetermined threshold value indicating the range to be obtained is obtained, it is determined to be non-living.

  FIG. 2A shows the luminance distribution of the face area A when a normal human face image is input, and FIG. 2B shows the case where a printed image having the face image is input. The luminance distribution of the face area B is shown. As is apparent from the figure, the luminance distribution varies depending on the difference in light reflectance and the distance from the position of the camera 101. Therefore, as shown in the graph of FIG. 2C, if the average luminance distribution of the face is in the range from the lower limit threshold TL to the upper limit threshold TU, it is determined as a normal person (living body), but FIG. When the average luminance is avg, as in the luminance distribution, the value is smaller than the lower limit threshold TL. In this case, it is determined that the input is not from a normal person but from a non-living body. In order to emphasize this difference, it is also possible to improve the discrimination accuracy by irradiating the auxiliary illumination light toward the face area. 2A shows the case of a normal person, and FIG. 3D shows the case of a printed matter.

  Similarly, when the face size is used, as shown in FIGS. 3A and 3B, the distances L1 and L2 between the eyes indicate the size of the face. Therefore, the distances L1 and L2 are normal. If it is outside the range of the size distribution when the person is imaged, it is determined that the input is from a non-biological body.

Further, when the contrast of the face area is used in the same manner, the contrast of the face areas A and B is caused by a blurring due to a difference in distance to the camera 101 as shown in FIGS. 3 (a) and 3 (c). Therefore, when the contrast falls outside the range of the contrast distribution when a normal person is imaged, it is determined that the input is from a non-biological body. As an index indicating contrast, for example, a differential image in the face areas A and B is created, and the average value of the differential values is averaged, or the image in the face areas A and B is Fourier transformed. It can also be realized by means for comparing the values of high-frequency components that are lost when the value decreases.
In this way, by comparing the image feature of the face area obtained from a normal person with the image feature of the input face area, it is possible to perform the biometric determination process without using a special device.

  Further, in the biometric discrimination unit 103, it is possible to improve discrimination accuracy by using not only image information but also a distance sensor. A living body discrimination device in the case where a distance sensor 105 as a distance measuring means is added to the first embodiment will be described as an example, and a configuration diagram thereof is shown in FIG.

  By simply using the information from the distance sensor 105, it is easy to determine that a living body is not present at a normal position, and further, from the camera 101 to the face Ma of the person M as shown in FIG. Even when the distance changes to d1 and d2, since the average face size S1 of the person M is known, the size of the face image inputted according to the distance can be calculated.

In other words, a case where a thing different from the face size of the person M is posted can be easily obtained from the ratio between the distance and the face size, and can be determined to be non-living when it falls outside the range of the predetermined ratio.
Similarly, if it is known that the luminance distribution also changes in advance according to the distance, a comparison can be made by obtaining the relationship between the distance information from the distance sensor 105 and the luminance information in advance, and the contrast It turns out that it is the same. In this case, since the means for reading biometric information cannot be used in the case of a contact type that comes in contact with the living body, it is difficult to realize with fingerprints, but the biometric information is not contacted such as the face, iris, retina, eyes, mouth, and veins. This is an effective means for acquisition.

  The output control unit 104 outputs information itself such as an input image input by the camera 101, a face area and feature information obtained from the face detection unit 102, and a discrimination result of the biometric discrimination unit 103 to an output destination device. The contents of the control signal are changed based on the discrimination result of the living body discrimination unit 103.

  In the simplest example, in an imaging device that simply records the obtained image or face area, it is possible to control the output so that no image is output from the output means when it is determined that the image is non-living. Similarly, a device capable of outputting voice and displaying a message is connected, and when it is determined that it is a living body, a greeting voice message is output, while when it is determined that it is a non-living body, it is controlled not to react. It is also possible. Conversely, an alarm may be generated in the case of a non-living body. In this case, in addition to outputting voice and messages with the device itself incorporating the biometric discrimination device, it is also possible to realize voice and message display to a remote device connected via a network or the like.

  Obviously, it is also possible to output a signal for controlling the connected device. For example, in the case of a traffic control device that permits passage when a face image that satisfies a predetermined condition is input, if it is determined that the face image is non-living, the biometric information is correctly output by not performing output to the traffic control device itself. It is possible to allow traffic only when input.

  Further, in the case of a program incorporating a living body discrimination means, the output result of the output control unit 104 corresponds to controlling whether or not to perform a predetermined process, and access to predetermined data managed by the program. It is possible to switch whether to allow or not.

  Further, it is possible to set the operation to be changed for each day of the week or time according to the determination result of the biometric determination unit 104 based on the table data as shown in FIG. When it is done, it is recorded in the history, but no alarm or notification is given immediately. If it is repeated many times, an alarm is given and the administrator is notified. On the other hand, it is also possible to easily realize the combined apparatus of the above-described embodiments, such as switching and managing alarms and notifications immediately at night and holidays when traffic is low.

  As described above, according to the first embodiment, a special biological reaction or operation is requested from a user in an apparatus that uses characteristic information or a state obtained by photographing biological information of a person. In addition, even when a photograph, a picture, a video display device, or the like exists within the shooting range, desired control can be performed without malfunction or misjudgment.

Next, a second embodiment will be described.
In the following description, an image is input by a camera as a person to be imaged (object from which biometric information is acquired), a face area is detected, and the object to be imaged is a real person (living body) or an object other than a person An example of a biological discrimination device that changes a control signal to a connected device by determining whether it is (non-living), but the present invention is not limited to a human face, It can be applied to those that perform biometric discrimination by inputting images containing various biometric information. For example, it can also be applied to biometric discrimination using iris information, retinal information, vein information of hands and fingers, eye and mouth status, etc. It is. For fingerprint information, when an optical sensor is used as an image acquisition unit, biometric discrimination can be performed according to the present embodiment.

  FIG. 7 schematically shows a configuration of a living body discrimination apparatus according to the second embodiment. The biometric discrimination device includes a camera 201 as an image acquisition unit that captures and inputs a face image (an image including at least a face) of a person (hereinafter, also referred to as a shooting target or an object) M to be shot. A face detection unit 202 serving as a biometric detection unit that detects the position of a face part such as a face, eyes, and nose from a face image of a person obtained from 201, and a subject to be imaged based on the image features of the face area detected by the face detection unit 202 The face image and face of the person M photographed by the living body discriminating unit 203 and the camera 201 as a living body discriminating means for discriminating whether M is a real living body or a printed matter such as a photograph or a picture or a video display device. Depending on the detection result of the face area detected by the detection unit 202 and the determination result of the biometric determination unit 203, the connected device, the operating output signal, and the program control means are changed. And an output control section 204 as an output control unit that.

  Since the camera 201, the face detection unit 202, and the output control unit 204 are the same as those in the first embodiment described above, the description thereof will be omitted, and the living body determination unit 203 that is different from the first embodiment will be described. .

  The biometric determination unit 203 detects whether there is an object other than a person such as a movement, a frame, a change in a background image, or an object that obstructs depth in addition to an area where biometric information exists, and determines whether it is a living body or a non-living body. Determine.

  First, a method for detecting that there is a movement outside the human body region will be described. In the present embodiment, since the face area is detected as the living body area, it is determined that the body is non-living when there is a change in the position away from the face area by a certain distance or more. For example, as shown in FIGS. 8A and 8B, if a normal person exists and the person does not move, there is no difference between the image one frame before and the current input image, and the difference between them. Finding the image shows no variation. In addition, even if a person moves, it can be assumed that fluctuations occur only near the person area.

  Therefore, as shown in FIG. 8C, when a position that is a predetermined distance away from the vicinity of the person outline C and the face area D is set as the background area E, and there is a variation in the difference image in the background area E Is determined to be non-living. Regarding the amount of variation, it is possible to calculate the total number of pixels that have changed in luminance by more than a predetermined value, measure the size of the area of the connected region, or detect that the pixels are arranged in a straight line. It is also possible to detect without detection. Since the face area is obtained by the face detection unit 202, the left area, the upper area, and the right area of the face area can be set as the background area, and the difference information of the background area can be easily detected. is there. For example, when a photograph or the like is posted, it is expected that not only the person portion but also the background area will change.

  Next, as another method, in a background area that is a predetermined distance away from the living body area detected by the face detection unit 202 and normally does not have an object, a rectangular shape, a curve, a polygon, or the like around the living body area When the object surrounding the living body region is detected, it is determined that the body is non-living.

  For example, for a photo with a face region as shown in FIG. 9A or a photo with a hand region as shown in FIG. 9B, the difference between the edge of the photo and the background is detected. Thus, the rectangle F surrounding the face area or the rectangle G surrounding the hand area can be detected. As the detection method, for example, line segment detection by Hough transform, vertical and horizontal projection processing for a linear component emphasized image by a Sobel filter can be detected, but other straight lines, closed curves, and polygons can be used. The same can be realized as long as the shape surrounding the living body region is detected by the detection method.

  Next, as yet another method, an image captured in a state where the person M does not exist is stored as a background image in advance, a difference image from the input image is obtained, and the biological region detected by the face detection unit 202 In a background area that is a predetermined distance away from the background image, it is determined that the background image is non-living when a pixel having a luminance change equal to or greater than a predetermined value is detected in a predetermined number or more or in a predetermined shape.

  For example, as shown in FIG. 10, when a person is present, the difference image as shown in FIG. 10B is obtained with respect to the image when no person is present in FIG. 10A, and a hatched range H is a change area from the background area. Become. Since the face area is detected here, the left, top, and right areas that are separated from the face area by a predetermined distance are the background areas, and there should be no change from the image when no person is present.

  Similarly, if a facial part such as an iris or retina is used, a similar comparison can be made by photographing a human face image at a wide angle, and the vein pattern of the hand is shown in FIG. It is possible to extract a hand shape such as a difference image. However, when any printed matter or video display device is posted as shown in FIG. 10D, the difference image is different from the shape of the person, and the background region also changes. By checking the pixels in this area, it is possible to determine whether it is a living body or a non-living body.

  Further, in the biometric discrimination unit 203, not only image information but also a distance sensor can be used together. A living body discrimination device when a distance sensor 205 as a distance measuring unit is added to the second embodiment will be described as an example, and a configuration diagram thereof is shown in FIG.

  The distance sensor 205 is a position where the person does not block the background when the normal person stands, that is, for example, as shown in FIG. 12, is away from the face area detected by the face detection unit 202, and Uses at least one position (K3 in this example) at positions K1, K2, and K3 corresponding to areas where no object exists.

  The biometric determination unit 203 compares each distance measurement value measured in the absence of a person with each distance measurement value measured in the presence of a person, and there is a difference of a predetermined value or more between both distance measurement values. Sometimes it is determined to be non-living. As the distance measurement points, the upper left part (K1), the upper right part (K3), the middle part (K2), etc. in the input image can be measured most easily.

  As described above, according to the second embodiment, a special biological reaction or operation is requested from a user in an apparatus that uses feature information or a state obtained by photographing biological information of a person. In addition, even when a photograph, a picture, a video display device, or the like exists within the shooting range, desired control can be performed without malfunction or misjudgment.

Next, a third embodiment will be described.
In the following description, an image is input by a camera as a person to be imaged (object from which biometric information is acquired), a face area is detected, and the object to be imaged is a real person (living body) or an object other than a person An example of a biological discrimination device that changes a control signal to a connected device by determining whether it is (non-living), but the present invention is not limited to a human face, It can be applied to those that perform biometric discrimination by inputting images containing various biological information. For example, iris information, retinal information, and eye and mouth states can be applied by acquiring a wide-angle image including the face area. In the case of using hand or finger vein information, it is difficult to apply when a capacitive contact sensor is used as the image acquisition means, but a non-contact type using an optical sensor or camera as the image acquisition means. When using The, it is possible to perform biometric determination by this embodiment.

  FIG. 13 schematically shows the configuration of the living body discrimination apparatus according to the third embodiment. This living body discrimination apparatus includes a camera 301 as an image acquisition unit that captures and inputs a face image (an image including at least a face) of a person (hereinafter, also referred to as a shooting target or an object) M to be shot. A face detection unit 302 serving as a living body detection unit that detects the position of a face part such as a face, eyes, and nose from a face image of a person obtained from 301, and a subject to be imaged based on the image features of the face area detected by the face detection unit 302 A biometric discrimination unit 303 as a biometric discrimination unit for discriminating whether M is a real living body or a printed matter such as a photograph or a picture or a video display device, and the face image or face of the person M photographed by the camera 301 Depending on the detection result of the face area detected by the detection unit 302 and the discrimination result of the living body discrimination unit 303, the connected device, the operating output signal, and the program control means are changed. Output control section 304 as an output control unit that is composed of an illumination unit 305 as an illuminating means capable of emitting illumination light to the living body region.

  Since the camera 301, the face detection unit 302, and the output control unit 304 are the same as those in the first embodiment described above, the description thereof will be omitted, and the illumination unit 305 and the living body determination unit different from those in the first embodiment. 203 will be described.

  The illumination unit 305 is configured by an illumination device that irradiates illumination light toward the face portion of the person M to be photographed. For example, various illuminations such as a normal incandescent lamp, a fluorescent lamp, a cold cathode tube, and an LED illumination lamp It can be realized with a device. In addition, depending on the living body to be imaged, it may be near infrared light or far infrared light instead of visible light. It doesn't matter if they match.

  The illumination unit 305 can be switched off and on at an arbitrary timing. The camera 301 acquires an image in a state of being turned off while being controlled by a living body determination unit 303 described later, and the image is displayed again even in the state of being turned on. It shall be controlled so that it can be obtained.

  The living body determination unit 303 controls the illumination unit 305 as described above to acquire an image irradiated with illumination light and an image not irradiated with illumination light, respectively. When the object M irradiated with the illumination light is a living body such as a person or a hand, the average luminance of the living body region should change by a predetermined value or more by turning on / off the illumination light because it is not an object that emits light by itself. It is.

  For example, as described in this embodiment, in the case of a face image, the image shown in FIG. 14A is when illumination light is irradiated, and the image shown in FIG. 14B is when illumination light is not irradiated. Thus, it can be seen that the face area when the illumination light is not irradiated becomes dark. On the other hand, when an image display device that emits light such as a CRT display or a liquid crystal display is present in the imaging range, the image display device is not significantly affected by the change in illumination light, and the predetermined display even when the illumination light is turned off. Since an image with luminance is displayed, the difference in luminance in the face area between when illumination light is irradiated and when it is not is reduced. FIGS. 14C and 14D show examples of images in which the change in luminance is small.

  In this way, it is possible to determine that the imaging target M is non-living when the brightness difference between the face area when irradiated with illumination light and when it is not irradiated is less than a predetermined value. It should be noted that even when the target object is a facial part, it can be realized by imaging the periphery of the facial part at a wide angle, and it is clear that the present invention can be applied even if the target object is a hand.

  As described above, according to the third embodiment, a special biological reaction or operation is requested from a user in an apparatus that uses characteristic information or a state obtained by photographing biological information of a person. In addition, even when a photograph, a picture, a video display device, or the like exists within the shooting range, desired control can be performed without malfunction or misjudgment.

Next, a fourth embodiment will be described.
In the following description, an image is input by a camera as a person to be imaged (object from which biometric information is acquired), a face area is detected, and the object to be imaged is a real person (living body) or an object other than a person An example of a biological discrimination device that changes a control signal to a connected device by determining whether it is (non-living), but the present invention is not limited to a human face, It can be applied to those that perform biometric discrimination by inputting images containing various biological information. For example, iris information, retinal information, and eye and mouth states can be applied by acquiring a wide-angle image including the face area. In the case of using hand or finger vein information, it is difficult to apply when a capacitive contact sensor is used as the image acquisition means, but a non-contact type using an optical sensor or camera as the image acquisition means. When using The, it is possible to perform biometric determination by this embodiment.

  FIG. 15 schematically shows a configuration of a living body discrimination apparatus according to the fourth embodiment. This living body discrimination device is an image acquisition unit that captures and inputs a face image (an image including at least a face) of a person (hereinafter also referred to as a shooting target or an object) M to be shot from a plurality of directions. A plurality of cameras 401, a face detection unit 402 as a biometric detection unit that detects the position of a face part such as a face, eyes, and nose from a face image of a person obtained from the camera 401, and a face area detected by the face detection unit 402 A person who is photographed by a living body discriminating unit 403 serving as a living body discriminating unit and a camera 401 that discriminates whether an imaging target M is a real living body or a printed matter such as a photograph or a picture or a video display device. Depending on the detection result of the face image detected by the face detection unit 402, the face area detected by the face detection unit 402, and the discrimination result of the living body discrimination unit 403, connected devices, operating output signals, and programs And an output control section 404 as an output control means for changing the control means of the beam.

  Note that the face detection unit 402 and the output control unit 404 are the same as those in the first embodiment described above, and thus the description thereof will be omitted, and only the camera 401 and the living body determination unit 403 that are different from those in the first embodiment will be described. .

  For example, as shown in FIG. 16A, the camera 401 includes two cameras 401a and 401b. The camera 401a is installed at a position for capturing a face image from the front of the person M, and the camera 401b has a direction other than the front. The position where the person M is photographed, that is, the angle at which the head of the person M is looked down from above the person M is installed. By installing the cameras 401a and 401b in this way, it is possible to photograph the face image of the front of the person M and the person image from above.

  The installation directions of the cameras 401a and 401b are not limited to the orientation shown in FIG. 16A, and the camera 401b other than the front of the person M can be installed at an angle other than the front such as the horizontal direction or the downward direction of the person M. is there.

  Further, for example, as shown in FIG. 16 (b), the camera 401b is installed in front of the person M on the upper part of the camera 401a, and a planar mirror 405 is provided on the upper part of the person M facing the camera 401b. By installing at an angle, images from different angles of the person M can be obtained. In this case, the living body discrimination device 400 including the cameras 401a and 401b only needs to be installed on the wall surface, and the mirror 405 can be installed at an arbitrary place, which facilitates installation.

  Further, for example, as shown in FIG. 16C, by using a curved mirror 406 as a mirror, it is possible to photograph human images from as many directions as possible. In the present embodiment, the description will be continued with an example in which the angle is set to look down from above.

  In addition, when the living body to be imaged is other than a face, for example, in the case of an iris, retina, eyes, or mouth, processing can be performed with an image similar to that for a face by enabling an image to be captured at a wide angle. . In the case of a fingerprint, the same can be realized by installing a plurality of optical sensors, and in the case of a hand vein pattern, the hand can be photographed from a plurality of directions.

  The living body determination unit 406 determines whether the imaging target M is a living body or a non-living body from the human images obtained at the above-described plurality of angles. For example, in each image obtained at a plurality of angles, a difference image with an image obtained when a person previously obtained is absent is created, and the results obtained by taking projection values in the vertical and horizontal directions are compared. Can be realized.

  As shown in FIG. 16A, when the camera 401a is installed in front of the person M and the camera 401b is installed at an angle looking down from above, the image of the living body region as shown in FIG. A vertical background differential projection value P11 and a horizontal background differential projection value P12 that match the shape are obtained, and the vertical direction that matches the shape of the living body region as shown in FIG. 17A with respect to the image from the camera 401b. The background difference projection value P21 and the background difference projection value P22 in the horizontal direction are obtained.

  On the other hand, when holding a printed matter or the like, a vertical background differential projection value P31 and a horizontal background differential projection value P32 as shown in FIG. 18A are obtained for the image from the camera 401a. A vertical background differential projection value P41 and a horizontal background differential projection value P42 as shown in FIG. 18A are obtained for the image from 401b. As is clear from these figures, the projected value of the living body region shown in FIG. 17 is greatly different, and indicators such as the maximum value, minimum value, average value, variance value, and area value of the projected value are compared. It is possible to discriminate whether it is a living body or a non-living body.

  In addition to this, a pattern for detecting a person's face is determined from the front direction, and a person pattern from the top of the head is obtained by learning in advance, and only when a person pattern is detected from any image, it is determined as a living body. This method can also be realized. Similarly, with respect to the hand region, taking the case of shooting from the front and side instead of the front and top, for example, the vertical background differential projection as shown in FIG. 19A with respect to the image from the camera 401a. A value P51 and a background difference projection value P52 in the horizontal direction are obtained, and a background difference projection value P61 in the vertical direction and a background difference projection value P62 in the horizontal direction as shown in FIG. 19A are obtained for the image from the camera 401b. Therefore, it is possible to realize distinction from printed materials and video display devices.

  As described above, according to the fourth embodiment, a special biological reaction or operation is requested from a user in an apparatus that uses characteristic information or a state obtained by photographing biological information of a person. In addition, even when a photograph, a picture, a video display device, or the like exists within the shooting range, desired control can be performed without malfunction or misjudgment.

1 is a block diagram schematically showing a configuration of a living body discrimination apparatus according to a first embodiment of the present invention. Explanatory drawing for demonstrating the change of the luminance distribution of a face area. Explanatory drawing for demonstrating the change of the face size of a face area. The block diagram which shows roughly the modification of the biometrics apparatus which concerns on 1st Embodiment. Explanatory drawing for demonstrating the relationship between the position of a biological body, and the magnitude | size of a biological body. Explanatory drawing for demonstrating the operation | movement table with respect to the operation | movement at the time of non-biometric image input. The block diagram which shows roughly the structure of the biometric discrimination apparatus which concerns on the 2nd Embodiment of this invention. Explanatory drawing for demonstrating the method to detect presence of objects other than the biological body using a difference image. Explanatory drawing for demonstrating the method to detect presence of objects other than a biological body using the shape detection of a biological body periphery. Explanatory drawing for demonstrating the method to detect presence of objects other than the biological body using a background difference. The block diagram which shows roughly the modification of the biometric discrimination apparatus which concerns on 2nd Embodiment. Explanatory drawing for demonstrating the method to detect presence of objects other than the biological body using a distance sensor. The block diagram which shows schematically the structure of the biometric discrimination apparatus which concerns on the 3rd Embodiment of this invention. Explanatory drawing for demonstrating the light-emitting body detection using on / off of illumination light. The block diagram which shows roughly the structure of the biometric discrimination apparatus which concerns on the 4th Embodiment of this invention. The schematic diagram for demonstrating the installation example of the camera with respect to several angles. The schematic diagram for demonstrating the example of an input image of a some angle. The schematic diagram for demonstrating the example of an input image of a some angle. The schematic diagram for demonstrating the example of an input image of a some angle.

Explanation of symbols

  M ... Shooting target, person, object 101, 201, 301, 401, 401a, 401b ... camera (image acquisition means), 102, 202, 302, 402 ... face detection unit (biological detection means), 103, 203, 303, 403 ... Living body discriminating unit (biological discriminating unit), 104, 204, 304, 404 ... Output control unit (output control unit), 105, 205 ... Distance measuring unit (distance measuring unit), 305 ... Illuminating unit (illuminating unit) ).

Claims (17)

A biological discrimination device for discriminating whether an object for acquiring biological information is a living body or a non-living body,
Image acquisition means for acquiring an image including biological information of the object;
A living body detecting means for detecting a living body region from the image acquired by the image acquiring means;
The image feature of the living body region detected by the living body detection means is compared with the image feature of the living body region obtained when a normal object obtained in advance is present at a predetermined position, A living body discriminating means for discriminating that the object is non-living when it is determined that there is a difference greater than
Based on the determination result of this biological determination means, output control means for changing the content to be output to the device connected as the output destination;
A living body discrimination device characterized by comprising: A biological discrimination device for discriminating whether an object for acquiring biological information is a living body or a non-living body,
Image acquisition means for acquiring an image including biological information of the object;
A living body detecting means for detecting a living body region from the image acquired by the image acquiring means;
Distance measuring means for measuring the distance from the image acquisition means to the object;
The relationship between the image feature of the living body area detected by the living body detecting means and the distance measured by the distance measuring means is the relationship between the image feature of the living body area obtained from the normal object obtained in advance and the normal object. Compared with the relationship between the object obtained when present at a predetermined position and the distance between the image acquisition means, and when it is determined that there is a difference greater than a predetermined value between the two objects, the object is not A living body discriminating means for discriminating that it is a living body,
Based on the determination result of this biological determination means, output control means for changing the content to be output to the device connected as the output destination;
A living body discrimination device characterized by comprising:   The biometric discrimination apparatus according to claim 1 or 2, wherein the image feature of the biometric region used in the biometric discrimination means is a size of the biometric feature.   The living body discrimination apparatus according to claim 1 or 2, wherein the image feature of the living body area used in the living body discrimination means is a luminance distribution of the living body area.   The living body discrimination apparatus according to claim 1 or 2, wherein the image feature of the living body area used in the living body discrimination means is a contrast of the living body area. A biological discrimination device for discriminating whether an object for acquiring biological information is a living body or a non-living body,
Image acquisition means for acquiring an image including biological information of the object;
A living body detecting means for detecting a living body region from the image acquired by the image acquiring means;
When it is detected whether an object other than a living body exists in an area other than the living body area detected by the living body detecting means in the image acquired by the image acquiring means, and the presence of an object other than the living body is detected And a living body discriminating means for discriminating that the object is non-living,
Based on the determination result of this biological determination means, output control means for changing the content to be output to the device connected as the output destination;
A living body discrimination device characterized by comprising:   The living body discriminating unit discriminates that the target object is non-living when a moving object is detected in a region that is a predetermined distance away from the living body region detected by the living body detecting unit and normally does not exist. The living body discrimination device according to claim 6.   The living body discriminating unit is a predetermined distance away from the living body region detected by the living body detecting unit, and usually in a region where there is no target, the rectangular region or a curve or a polygon around the living body region. The living body discriminating apparatus according to claim 6, wherein when the object surrounding the living body area is detected, the target object is determined to be a non-living body. A biological discrimination device for discriminating whether an object for acquiring biological information is a living body or a non-living body,
Image acquisition means for acquiring an image including biological information of the object;
A living body detecting means for detecting a living body region from the image acquired by the image acquiring means;
By obtaining a difference image between an image acquired by the image acquisition unit and an image acquired in advance in the absence of an object, a predetermined distance from the biological region detected by the biological detection unit is determined. A living body discriminating means for discriminating that the object is non-living when an area having a luminance change of more than a value is detected in a predetermined number or more or in a predetermined shape;
Based on the determination result of this biological determination means, output control means for changing the content to be output to the device connected as the output destination;
A living body discrimination device characterized by comprising: A biological discrimination device for discriminating whether an object for acquiring biological information is a living body or a non-living body,
Image acquisition means for acquiring an image including biological information of the object;
A living body detecting means for detecting a living body region from the image acquired by the image acquiring means;
A distance measuring unit that is a predetermined distance away from the living body region detected by the living body detecting unit, and that normally corresponds to a region where no object exists,
The distance measured by the distance measuring means in the absence of the object is compared with the distance measured in the presence of the object, and when there is a difference of a predetermined value or more between the distances, the object Is a living body discriminating means for discriminating that it is non-living,
Based on the determination result of this biological determination means, output control means for changing the content to be output to the device connected as the output destination;
A living body discrimination device characterized by comprising: A biological discrimination device for discriminating whether an object for acquiring biological information is a living body or a non-living body,
Image acquisition means for capturing and acquiring an image including biological information of the object;
Illuminating means for irradiating the object with illumination light during photographing by the image obtaining means;
A living body detecting means for detecting a living body region from the image acquired by the image acquiring means;
Luminance distribution in a living body region detected by the living body detecting unit with respect to an image acquired by the image acquiring unit in a state in which illumination light is irradiated by the illuminating unit and an image acquired in a state in which illumination light is not irradiated by the illuminating unit. A living body discriminating means for discriminating that the object is non-living when the difference between the obtained luminance distributions is less than a predetermined value;
Based on the determination result of this biological determination means, output control means for changing the content to be output to the device connected as the output destination;
A living body discrimination device characterized by comprising: A biological discrimination device for discriminating whether an object for acquiring biological information is a living body or a non-living body,
Image acquisition means for acquiring an image including biological information of an object from a plurality of directions;
A living body detecting means for detecting a living body region from each of the plurality of images acquired by the image acquiring means;
By this living body detection means, it is confirmed whether or not the living body area has been detected from all of the plurality of images acquired by the image acquiring means, or whether or not the shape of the detected living body area is a predetermined shape, It is determined that the target object is a living body only when the living body area is detected from all of a plurality of images, or when the shape of the detected living body area is confirmed to be a predetermined shape. A biological discrimination means for discriminating that the object is non-living,
Based on the determination result of this biological determination means, output control means for changing the content to be output to the device connected as the output destination;
A living body discrimination device characterized by comprising:   The living body discrimination apparatus according to claim 12, wherein the image acquisition unit acquires images from a plurality of directions by a plurality of imaging units installed in a plurality of different directions.   A biological discrimination method for discriminating whether an object from which biological information is acquired is a living body or a non-living body, an image including biological information of the target is acquired, and a biological region is detected from the acquired image The image feature of the detected living body region is compared with the image feature of the living body region obtained when a normal object obtained in advance is present at a predetermined position. A living body discrimination method characterized in that when it is determined that there is a difference, the object is determined to be non-living.   A biological discrimination method for discriminating whether an object from which biological information is acquired is a living body or a non-living body, an image including biological information of the target is acquired, and a biological region is detected from the acquired image In the acquired image, it is detected whether an object other than the living body is present in an area other than the detected living body area, and when the presence of an object other than the living body is detected, the target object is a non-living object. A biometric discrimination method characterized in that it is discriminated.   A biological discrimination method for discriminating whether a target for acquiring biological information is a living body or a non-living body, and acquires an image including biological information of the target in a state where illumination light is irradiated on the target. In addition, an image including biological information of the target object is acquired in a state where the target object is not irradiated with illumination light, a biological area is detected from each of the acquired images, and a luminance distribution in the detected biological area is calculated. A living body discrimination method characterized by determining each object and determining that the object is non-living when a difference between the obtained luminance distributions is less than a predetermined value.   A biological discrimination method for discriminating whether an object from which biological information is acquired is a living body or a non-living body, an image including biological information of the object is acquired from a plurality of directions, and the plurality of acquired images And detecting whether or not the living body area is detected from all of the plurality of images, or whether or not the detected living body area is a predetermined shape. It is determined that the target object is a living body only when the living body area is detected from all the images or when it is confirmed that the shape of the detected living body area is a predetermined shape. Is a living body discrimination method characterized by discriminating that the object is non-living.

Images