JPH0721366A - Individual identifying device - Google Patents

Abstract

PURPOSE:To provide an individual identifying device with improved safety and less burdens of a user utilizing the information of a living body for which feature points are less compared to fingerprints. CONSTITUTION:This device is constituted of an illumination means constituted of a light source 3 and a light projecting lens group 4, a prism 5 as a transparent body block and an image reception means composed of an image forming lens group 9 and an image reception element 10 and is provided with an input device 2 for placing a hand 1 and fetching the images of a palm as pictures and a computer 14 for processing picture data outputted from the image reception element 10. The computer 14 demonstrates a function as a storage means for storing a feature amount as individual identification information at the time of registration and also functions as a discrimination means for judging whether or not the stored feature amount coincides with the feature amount extracted from palm images obtained from the output of the image reception element 10 at the time of collation.

Description

Detailed Description of the Invention

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a personal identification device which constitutes, for example, a part of a security system, and more particularly to a device which uses individual differences of living bodies as identification information.

[0002]

2. Description of the Related Art Conventionally, personal identification devices used for entry / exit management of security systems include a method of inputting a personal identification number, a method of magnetically or optically reading information on a card in which an ID number is registered, and a fingerprint. There are known methods using biometric information such as, iris, and fundus blood vessels.

For example, a device utilizing a fingerprint is disclosed in Japanese Patent Laid-Open No.
A system utilizing an iris is disclosed in Japanese Patent Application Laid-Open No. 62-501889.

[0004]

However, in the method of using a personal identification number, there is a possibility that the personal identification number will be leaked to another person due to threats or the like, and in the method of using a card, the card can be lost or stolen to another person's hand. There is a problem with safety.

Further, the method using the iris and the fundus blood vessel is higher in safety than the above-mentioned method, but has a problem that it is necessary to illuminate the eyes for detection, which imposes a heavy burden on the user.

Further, since the fingerprint has a complicated pattern, the number of characteristic points is 100 or more, and a method using the fingerprint requires a lot of time for the identification process.

[0007]

SUMMARY OF THE INVENTION The present invention has been made in view of the above-mentioned problems of the prior art, and provides biometric information with high safety, less burden on the user, and less characteristic points than fingerprints. It is intended to provide a personal identification device used.

[0008]

Means for Solving the Problems The palm is composed of a raised portion formed of thick skin and a wrinkled portion of thin skin. The palm has three relatively thick and long wrinkles generally called fate line, lifeline, and brain line, and the palm that is the pattern of these wrinkles is different for each individual, so use this as personal identification information. You can

The personal identification device according to the present invention is characterized in that this palm is used as identification information. Specifically, the input means for capturing an image of the palm as an image and the feature amount of the individual from the captured palm image. An extraction unit for extracting the feature amount, a storage unit for storing the feature amount as personal identification information at the time of registration, and a feature amount stored in the storage unit for matching.
And a discriminating unit for discriminating whether or not the feature amount extracted based on the palm image output from the input unit matches.

[0010]

Embodiments of the personal identification device according to the present invention will be described below. FIG. 1 shows a basic configuration example of the personal identification device of the embodiment.

As shown in FIGS. 1 and 2, this apparatus includes an input device 2 on which a hand 1 is placed and a palm image is captured as an image. The input device 2 includes an illuminating unit including a light source 3 and a light projecting lens group 4, a prism 5 as a transparent block, and an image receiving unit including an image forming lens group 9 and an image receiving element 10. There is.

The light source 3 is composed of a fluorescent lamp, an incandescent lamp, a light emitting diode, electronic luminescence, a laser and the like. The prism 5 includes an end face 6 on which the illumination light is incident and a detection face 7 that is a curved surface that is convex on the side where the hand 1 is placed.
And an end face 8 from which reflected light is emitted. A CCD sensor or the like is used for the image receiving element 10.

Illumination light emitted from the light source 3 and incident on the prism 5 from the end face 6 through the light projecting lens group 4 is internally reflected by the end face 8 and enters the detection surface 7 from directly below. The illumination light that reaches the detection surface is partially reflected by the detection surface because a space is created between the detection surface and the palm at the portion corresponding to the wrinkle, and the remaining portion is emitted to the outside of the prism to the raised portion. At the corresponding part, the palm is in close contact with the detection surface and is scattered.

In this example, the image forming lens group 9 and the image receiving element 10, which are image receiving means, are arranged in such a range that only light rays reflected at angles larger than the critical angle reach all points on the detection surface. The reflected light from the wrinkles does not enter the image receiving means, but only the scattered light from the ridges emitted from the end face 8 enters. Therefore, the resulting image has a black wrinkled image in the white ridge.

Reference numeral 11 in the figure is a switch for instructing the start of palm image input. The switch 11 is provided at a position touched with a fingertip when the palm is placed on the detection surface 7. The switch 11 may be a pushbutton switch in the simplest case, or the contact of the fingertip is detected photoelectrically, or the capacitance of the capacitance is changed. It may be a switch that is regarded as a change. Further, the position where the switch is provided is not limited to this embodiment, and the input of an ID number, which will be described later, can be used as a signal for starting the input without providing the switch.

The image data output from the image receiving element 10 is displayed on the monitor 12 and stored in the image input device 13 having a frame memory. The computer 14 reads the image data stored in the image input device 13 as needed, and forms an image of the palm pattern to extract an individual feature amount.

The computer 14 registers at the time of registration.
A function as a storage unit that stores the feature amount as personal identification information is exhibited, and at the time of matching, the stored feature amount and the feature amount extracted from the palm phase image obtained from the output of the image receiving element 10 match? It functions as a determination means for determining whether or not.

The monitor 11 is mainly used for confirmation at the time of registration, and does not need to be provided when it is used as a device dedicated to collation. Further, the display device 15 is used for displaying various operation procedures at the time of registration, displaying operation confirmation information at the time of matching, and the like.

When the above system is used for the entrance / exit management of the security system, at least the input device 2 is arranged in the vicinity of the door provided with the automatic lock to be controlled, and this lock is commanded from the computer 14. Controlled by.

When there are a plurality of doors to be controlled, one input device 2 is provided for each door as shown in FIG. 3, and the image input device 13 and the computer 14 are provided.
It is also possible to provide one set of and to perform centralized management.

Furthermore, it is possible to connect a plurality of unit systems with the system of FIG. 3 as a unit to form a complex system. FIG. 4 shows an example of the complex system, in which computers 14 of each unit system are connected by a communication network, and information such as individual identification data, unlocking history, or usage status of each unit system is exchanged. can do.

Such a complex system can be used, for example, when facilities of the same organization are distributed in a plurality of places,
By using the identification data registered in one facility in other facilities as well, it is not necessary to individually register in each facility, and efficient operation is possible.

Next, the operation of the apparatus of the embodiment will be described with reference to FIG. In the example of FIG. 5, (a) registration and (b) verification are processed by separate programs, but these two functions may be processed by a common program.

When the process starts, in step 1 (indicated by "S." in the drawing), an input of an ID number for personal identification is prompted, and when this is input, in step 2, the image data of the palm is read from the image sensor 10. Is entered. The process may be started by the operator starting the registration program, or by using the ID number input in step 1 as a trigger.

The ID number may be directly input using a keyboard, or may be input by placing a magnetic card or the like on a reading device.

In step 2, when the hand is placed at a predetermined position, the switch 11 detects this, the light source 3 is turned on, and the image is input to the image input device 13. Alternatively, the light source 3 may be turned on by inputting the ID number. When the image input is completed, a notification means (not shown) informs the person to be measured that the image input is completed by voice, light or the like. At the time of registration, it is possible to confirm whether or not a good image is obtained while observing the display on the monitor 12, and repeat the image input until a good image is obtained.

In step 3, the image data stored in the image input device 13 is processed in the computer 14 to improve the image quality. As shown in FIG. 6A, the input image data includes three relatively thick wrinkles 16 and 1 to be detected.
In addition to 7 and 18, fine wrinkles that become noise are also included. From this, the image data is binarized in order to remove fine wrinkles and noise, and the image shown in FIG. 6B is taken out. Subsequently, image expansion / contraction processing is performed to obtain image data as shown in FIG. It should be noted that due to individual differences, fine wrinkle information may remain even after image quality improvement processing is performed, but wrinkles that are not erased even by image quality improvement have meaning as identification information for identifying an individual, and therefore are not necessarily shown in the figure. It is not necessary to extract only three wrinkles as shown in 6 (c).

Next, in step 4, in order to extract the feature quantity, the feature points as shown in (1) to (6) in FIG. 7 are picked up by a line tracing method or the like. The feature points are
There is an end point as in (1) and a branch point as in (2), and numbers are given in order from the center closer to the center of the concentric circle displayed on the circular screen.

As the feature quantity, the number and positional coordinates of each feature point indicated by black dots in FIG. 7, the coordinates of points obtained by equally dividing the branch line extending from each feature point as indicated by * in FIG. The directional vector of the branch line indicated by the arrow in FIG. 9 and the connection relation between the feature points, the angle formed by each vector indicated by the symbols a to d in FIG. 9, and the branch line indicated by e to i in FIG. An angle (h = 0 in this example) formed by a straight line connecting the division points can be considered. These feature quantities are used alone or in combination as identification information for identifying an individual.

At the time of registration, the process proceeds from step 4 to step 5, the extracted characteristic amount is recorded together with the ID number in the disk device or the like to save the data, and the process is ended.

At the time of matching, the feature amount is called from the stored data registered in step 6 using the ID number as a key, and the feature amount extracted from the data input in step 7 and the feature amount of the stored data are compared. To compare.
Since it is predicted that the direction of the palm will rotate during registration and during verification, a predetermined allowable range may be set during comparison. When the angle formed by the vector or the connection relation between the feature points is used as the feature amount, the influence of rotation can be reduced.

Further, when the above deviation exceeds the permissible range or in response to an erroneous operation such as an ID number input error, the operation from step 1 is repeated a predetermined number of times even when the comparison result is another person at the time of collation. You may set it.

As a result of the comparison, if the person is confirmed to be the person,
When the process proceeds from step 8 to step 9 and the corresponding action such as unlocking the door is completed and the process is terminated, and if the person cannot be confirmed, the process proceeds to step 10 and a denial action such as not releasing the lock is performed. Ends the process.

Instead of saving the image data as it is,
Since the feature quantity is extracted and then saved, the information quantity of the individual data to be saved can be made compact, and
It is possible to quickly determine even at the time of collation.

In the above example, the ID number and the feature amount of the palm are linked and managed. However, when the matching process time is short, that is, the number of registered data items is small or When the data amount of the feature amount is limited to a small amount, the feature amount extracted from the input image is compared with the registered feature amount data without using the ID number. By doing so, it is possible to collate.

Next, another configuration example of the input device 2 described above will be described. Regarding the arrangement of the illuminating device and the imaging device with respect to the prism 5, the following three combinations can be considered depending on the way of reflecting the illumination light on the detection surface 7.

The first is a case where the total reflection light on the detection surface is directly used. In this case, the illumination light is made incident on the detection surface at an angle larger than the critical angle at each point on the detection surface, and the image receiving means is arranged in an angle range larger than the critical angle.
The illumination light is totally reflected by the detection surface at the portion corresponding to the wrinkle and reaches the image receiving means, and is scattered at the portion corresponding to the raised portion and a part thereof reaches the image receiving means. Therefore, the resulting image is white and wrinkled in the gray ridge.

The second is a case where the total reflection light on the detection surface is indirectly used. In this case, the illumination light is made incident on the detection surface at an angle larger than the critical angle, and the image receiving means is arranged in a range where the totally reflected light does not reach. The totally reflected light does not reach the angle range smaller than the critical angle on the side where the totally reflected light flux exits, and the same side as the illumination means. Illumination light is totally reflected on the detection surface at the portion corresponding to the wrinkle and does not reach the image receiving means, but a part of the scattered light at the portion corresponding to the raised portion reaches the image receiving means. Thus, the resulting image will have a black wrinkled image within the gray ridge. However, when the image pickup means is provided on the same side as the illuminating means, it is difficult to increase the contrast of the image because the light returning from the raised portion is relatively small.

Third, there is no use of totally reflected light on the detection surface. In this case, the illumination light is made incident on the detection surface at an angle smaller than the critical angle, and the image receiving means is arranged in an angle range larger than the critical angle. In the part corresponding to the wrinkle, the illumination light partially passes through the detection surface and reaches the wrinkle, is reflected by the wrinkle and enters the detection surface, and the rest is reflected by the detection surface. The image receiving means is not reached because it proceeds inside. On the other hand, a part of the scattered light at the portion corresponding to the raised portion reaches the image receiving means. Therefore, the obtained image has a black wrinkled image formed in the white ridge.
In the configuration shown in FIG. 1 described above, the devices are arranged so as to satisfy the third condition.

In the second embodiment shown in FIG. 11, the above-mentioned first
Each means is arranged under the condition of. The illumination light emitted from the light source 3 and incident on the prism 5 from the end face 6 through the light projecting lens group 4 is incident on the detection surface 7 at an angle larger than the critical angle, so that the portion corresponding to the wrinkle portion is totally reflected. At the same time, it is scattered at the portion corresponding to the raised portion. Part of the total reflected light from the wrinkles and the scattered light at the raised portions exits from the end face 8 and forms an image on the image receiving element 10 via the imaging lens group 9.

In the third embodiment shown in FIG. 12, each means is arranged under the above-mentioned second condition. The illumination light emitted from the light source 3 and transmitted through the light projecting lens group 4 enters the prism 5 from the end surface 8 of the prism. Since the illumination light is incident on the detection surface 7 at an angle larger than the critical angle, the illumination light is totally reflected at the portion corresponding to the wrinkle and is emitted from the other end surface 6 of the prism to the outside of the prism. A part of the illumination light scattered at the portion corresponding to the raised portion is emitted from the end face 8 to the outside of the prism and reaches the image receiving element 10 via the imaging lens group 9.

In the fourth embodiment shown in FIG. 13, the respective means are arranged under the above third condition. Illumination light emitted from the light source 3 and transmitted through the light projecting lens group 4 enters from the end surface 6 of the prism 5 and enters the detection surface 7 at an angle smaller than the critical angle. A part of the illumination light is reflected by the detection surface at the portion corresponding to the wrinkle, the rest is emitted outside the prism, and is scattered at the portion corresponding to the raised portion.

Since the image forming lens group 9 and the image receiving element 10 which are image receiving means are arranged in an angle range larger than the critical angle, the reflected light from the wrinkles does not enter the image receiving means and the end face 8 is formed. Only the scattered light emitted from the raised portion is incident.

In order to satisfy the third condition, the light source 3 and the light projecting lens group 4 may be arranged so that the illumination light enters from the end face 8 side as shown by the broken line in FIG. The light source shown by and the light source shown by the broken line may be turned on at the same time. Further, as in the fifth embodiment shown in FIG. 14, illumination may be performed from directly below the prism 5. However, when the illumination means is provided at the position shown by the broken line in FIG. 13 or the position shown in FIG.
It is necessary to take care so that the light beam directly reflected by does not reach the image receiving means.

The sixth embodiment shown in FIG. 15 and the seventh embodiment shown in FIG. 16 are both arranged to satisfy the above condition 3. The illumination light that has entered from the bottom surface 6 enters perpendicularly to the tangent plane of the apex of the detection surface 7, and the scattered light from the raised portion exits from the end surface 8 and reaches the image receiving element 10.

The shape of the prism 5 as the transparent block is not limited to a triangular prism, but may be a trapezoid, a pyramid, a truncated pyramid, or the like if the conditions are satisfied. Further, the curved surface shape of the detection surface of the prism may be not only a spherical surface but also a rotationally symmetric aspherical surface, an ellipsoidal surface, or a rotationally asymmetrical aspherical surface.

The height difference h (see FIG. 17) from the peripheral portion of the detection surface 7 to the apex may be set according to the depression of the palm, but it depends on the individual difference and the force applied to the palm. A variation of about 1 mm to 10 mm was recognized depending on the insertion condition.
(30 samples). Therefore, h is 2 mm to 8 m
If it is set in the range of m, the shape generally fits the palm. The diameter w of the curved surface portion of the detection surface may be in the range of 45 mm to 75 mm.

[0048]

As described above, according to the present invention, by using a palm as identification information for identifying an individual, it is safer and less burdens the user, and more than using a fingerprint. It is possible to provide a personal identification device that can be discriminated with a small number of characteristic points, that is, in a short time.

Further, by using a transparent block having a convex curved surface as a detection surface as the input means, the image of the entire palm can be accurately detected.

[Brief description of drawings]

FIG. 1 is an explanatory diagram showing a first embodiment of a personal identification device according to the present invention.

FIG. 2 is a perspective view of the input device shown in FIG.

FIG. 3 is a system diagram of a unit system when a plurality of input devices are managed by one computer.

FIG. 4 is a system diagram of a complex system in which one computer or a plurality of computers that manage a plurality of input devices are connected by a communication network.

FIG. 5 is a flowchart showing the operation of the apparatus according to the embodiment of the present invention.

6A is an explanatory diagram showing an image of an input palm, FIG.
(b) is an explanatory view showing an image obtained by binarizing the image data of (a), and (c) is an explanatory diagram showing an image obtained by expanding and contracting the image data of (b).

7 is an explanatory diagram showing a state in which feature points are extracted from the image data of FIG.

FIG. 8 is an explanatory diagram of a case where a feature amount is regarded as coordinates of division points between feature points.

FIG. 9 is an explanatory diagram in a case where the feature amount is regarded as a vector connecting the dividing points of the feature points or a vector connecting the feature points, or an angle formed by these vectors.

FIG. 10 is an explanatory diagram of a case where a feature amount is regarded as an angle formed by a straight line connecting division points.

FIG. 11 is an explanatory diagram showing an optical system of the input device according to the second embodiment.

FIG. 12 is an explanatory diagram showing an optical system of an input device according to a third embodiment.

FIG. 13 is an explanatory diagram showing an optical system of an input device according to a fourth embodiment.

FIG. 14 is an explanatory diagram showing an optical system of an input device according to a fifth embodiment.

FIG. 15 is an explanatory diagram showing an optical system of an input device according to a sixth embodiment.

FIG. 16 is an explanatory diagram showing an optical system of an input device according to a seventh embodiment.

FIG. 17 is an explanatory diagram showing the size of the curved surface of the prism.

[Explanation of symbols]

 DESCRIPTION OF SYMBOLS 1 ... Hand 2 ... Input device 3 ... Light source 4 ... Projection lens group 5 ... Prism 7 ... Detection surface 6 ... End surface 8 ... End surface 9 ... Imaging lens group 10 ... Image receiving element 14 ... Computer

Claims (5)

[Claims] 1. Input means for capturing a palm image as an image, extracting means for extracting a characteristic amount of an individual from the captured palm image, and storage means for storing the characteristic amount as personal identification information at the time of registration. At the time of matching, it is provided with a determination unit that determines whether or not the feature amount stored in the storage unit and the feature amount extracted based on the palm image output from the input unit match. Personal identification device. 2. The input device includes a transparent block having a curved surface for closely contacting the palm, an illuminating device for illuminating the palm from the inside of the transparent block, and a reflected light from the curved surface for close contact with the palm. The personal identification device according to claim 1, comprising an imaging device that receives light and detects an image of a palm. 3. The personal identification device according to claim 1, wherein each of the input unit, the extraction unit, the storage unit, and the determination unit is provided. 4. A plurality of the input means are provided, and one each of the extraction means, the storage means, and the determination means are provided, and the information from the plurality of input means is collectively processed. The personal identification device according to claim 1. 5. A plurality of said input means, said extraction means, said storage means, and said discriminating means are provided respectively, and a communication network capable of data transfer between said plural discriminating means is provided. The personal identification device according to claim 1.