JPH1063858A - Method for identifying individual and device therefor - Google Patents

Abstract

PROBLEM TO BE SOLVED: To attain individual identification with high precision by obtaining sufficient information amounts even when the level of the opening of eyes is relatively small. SOLUTION: Iris data 3 obtained by extracting the characteristics of iris, and white data 5 obtained by extracting the characteristics of the white of the eyes surrounding the iris are registered with a registering date 7. At the time of identifying an individual, the similarity of the iris and the similarity of the white data are altogether judged. Weighting is operated to the iris similarity and the white data similarity, and when a long time passes after the registration, the white data are collected and registered again.

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a personal identification method and an individual identification device for confirming the identity of a person by using an eye image when entering or leaving a room or performing various transactions.

[0002]

2. Description of the Related Art Among the images of the eyes, in particular, the image of the iris is used as information for identifying an individual similarly to a fingerprint (Japanese Utility Model Laid-Open No. 3-101911, Japanese Patent Publication No. 5-84166). Gazette). These iris data are
The information is previously collected from a room or building that is required to enter or pass through a room or building that requires confidentiality, and is registered in the storage device. Then, the data is compared and collated with the data acquired each time the room is entered.

[0003]

However, the above-described conventional personal identification method has the following problems to be solved. The color of the iris differs depending on the race, and the method of obtaining it is also selected. Further, depending on the shape of the eyes and the like, there are those who easily acquire the iris and those who have difficulty acquiring the iris. Old Japanese people may not be able to obtain enough information on iris data even if they take pictures of the eyes when trying to enter a room while standing in front of a door, for example, because the age of the eyes is small. . Therefore, it becomes impossible to determine whether or not the user is himself / herself by comparing with the iris data collected in advance, and a problem that entry into the room is not permitted occurs. In this case, a system for personal identification using iris data cannot be effectively used.

[0004]

The present invention employs the following structure to solve the above problems. <Structure 1> The iris data from which the features of the iris are extracted and the white eye data from which the features of the surrounding white eyes are extracted and registered in advance, and the iris data obtained for personal identification and the registered iris data , Calculating the similarity of the iris data, comparing the white eye data acquired for personal identification with the registered white eye data, calculating the similarity of the white eye data, and calculating the similarity of the iris data. A personal identification method characterized in that, when the similarity between the iris data and the white-eye data is at a certain level or higher, the iris data and the white-eye data acquired for the personal identification are determined to belong to the individual.

<Explanation> Data for personal identification is acquired each time someone passes through a gate or the like. Similarity is the degree to which the content of the data matches,
The calculation method is arbitrary. Whether it is at or above a certain level can be determined by providing a threshold value or the like. The threshold for iris data and the threshold for white-eye data may be different. Further, the similarity may be added to or multiplied by a predetermined arithmetic expression, and the overall judgment may be made from the result.

<Structure 2> In structure 1, the total similarity is obtained by adding the similarity of the iris data and the similarity of the white-eye data, and when the total similarity is at a certain level or more, the personal identification is performed. Therefore, it is assumed that the acquired iris data and the white eye data are determined to be those of the individual, and the weighting coefficient is applied so that the similarity of the iris data has a greater effect on the overall similarity than the similarity of the white eye data. Characteristic personal identification method.

<Explanation> A weighting coefficient may be added to iris data, white eye data, or both. In any case, the similarity of the iris data may have a greater effect on the overall similarity than the similarity of the white-eye data. The weighting is performed in this manner in order to use the white-eye data in an auxiliary manner.

<Structure 3> A personal identification method according to Structure 2, wherein a weighting coefficient is selected such that a rate of affecting the overall similarity of the white-eye data gradually decreases with time.

<Explanation> When the white-eye data obtained from the same person tends to fluctuate with the passage of time, the influence on the individual identifiability is gradually reduced with the passage of time to prevent erroneous recognition.

<Structure 4> A personal identification method according to Structures 1 to 3, wherein white eye data is newly collected and re-registered when a predetermined time has elapsed.

<Explanation> When the white-eye data obtained from the same person may change after a long period of time,
Re-registered to prevent false recognition.

<Structure 5> In the structure 1, when the shape of the eye photographed for personal identification is large enough to acquire a sufficient amount of iris data, the similarity of only the iris data is used for personal identification. Therefore, the obtained iris data is determined to be that of the individual, and if the shape of the eye is too large to obtain a sufficient amount of iris data, the similarity between the iris data and the white-eye data is used for personal identification. A personal identification method, wherein the acquired iris data and white eye data are determined to belong to the person.

<Explanation> When it is determined that it is sufficient to acquire iris data from an image of an eye photographed by a camera or the like, personal identification can be performed only with the iris data, so that the processing for white eye data is omitted. In this way, the white-eye data can be appropriately used as an aid when personal identification is not easy using only the iris data.

<Structure 6> A registered data storage unit for registering and storing iris data from which the features of the iris are extracted and white eye data from which the features of the white of the eye around the iris are extracted, and photographing the eyes for personal identification. Receiving the outputs of the iris data acquisition unit for acquiring iris data, the white eye data acquisition unit for acquiring white eye data, the iris data acquisition unit and the white eye data acquisition unit,
The similarity between the iris data and the white-eye data is calculated by comparing with the corresponding data in the registration data storage unit, and when the similarity between the iris data and the white-eye data is at a certain level or more. A personal identification device, comprising: a matching unit that determines that the iris data and the white-eye data acquired for personal identification belong to the user.

<Description> The format of the registration data is arbitrary.
The iris data acquisition unit and the white-eye data acquisition unit may be separate bodies, integrated, or partially common.

<Structure 7> In the structure 6, the iris data, the white-eye data, and the registration date of the white-eye data are registered in the registration data storage unit, and the matching unit checks the similarity of the iris data and the similarity of the white-eye data. , And when the overall similarity is at or above a certain level, the iris data and the white-eye data acquired for personal identification are determined to be those of the individual, and the iris data So that the degree of similarity has a greater effect on the overall similarity than the degree of similarity of the white-eye data,
An elapsed time calculation unit that applies a weighting coefficient and selects a weighting coefficient so that the rate of affecting the overall similarity of the white-eye data gradually decreases with time, and calculates the time elapsed from the registration date and time. A personal identification device characterized by the following.

<Description> The registration date and time may be the date and time actually stored in the registration data storage unit, the date and time when the white-eye data was collected, or the date and time in the vicinity thereof. This is because it is negligible for calculating the elapsed time. The date and time may include only the day or the time. Further, when the white-eye data is stable for a long period of time, the month may be managed because the month may be managed.

<Structure 8> In the structure 6 or 7, the camera uses a different illumination with a different focal point when photographing an iris image and when photographing a white-eye image. Identification device.

<Explanation> In the case of Japanese, the iris is brown, and the white eye includes a pattern of red capillaries on a white background, and it is preferable to change the color of illumination for photographing. Further, in order to shoot a precise image, it is preferable to change the focal length between the case of shooting an iris and the case of shooting a white eye. As a result, highly reliable data for personal identification can be obtained.

[0020]

DESCRIPTION OF THE PREFERRED EMBODIMENTS Embodiments of the present invention will be described below using specific examples. <Personal Identification Method> FIG. 1 is an explanatory diagram showing a specific example of the personal identification method of the present invention. In the present invention, white eye data is used in addition to iris data. Prior to the description of FIG. 1, a method of acquiring iris data and white-eye data will be described in order. As described in the prior art document already introduced, when acquiring iris data, an eye is photographed in advance, and the pupil center is obtained to organize the data.

FIG. 2 is a diagram for explaining a method of determining the center of the pupil. When the eyes are photographed and the outline of the iris 11 is recognized, FIG.
As shown in (2), two tangents 12, 14 are drawn on the contour. Then, by drawing straight lines 13 and 15 perpendicular to this tangent line, the intersection is defined as the pupil center 9. The contour of the iris 11 can be obtained by capturing a portion where the brightness changes from the image of the eye photographed using a CCD camera or the like and cutting out the image.

Subsequently, preparations for obtaining iris data are performed. FIG. 3 is an explanatory diagram of the setting contents of the analysis band of the iris. As shown in this figure, an arbitrary number of analysis bands 16 on concentric circles are set from the pupil center 9 to the periphery of the iris 11.
Then, for each analysis band 16, the change in shading is digitally coded.

FIG. 4 is a diagram for explaining an example of digital coding of light and shade. The upper half of this figure is analog data, and the lower half is digital data. The vertical axis represents shading, and the horizontal axis represents the rotation angle θ with respect to the radius vector r shown in FIG. That is, one of the analysis bands 16 is traced in the circumferential direction, and the change in light and shade is graphed as shown in this figure, a threshold value K is set. Digitized as follows. Thus, iris data is obtained.

FIG. 5 is a diagram for explaining the relationship between the eye image and the iris data. If the iris data as described above is collected from each person to be identified and registered in advance, for example, the eyes of a person passing through the gate are photographed and the iris data is acquired. Identification is possible. In FIG.
FIG. 2 is a diagram illustrating the relationship between an eye image and iris data. However, as shown in FIG. 5A, when the eye 10 is sufficiently open and the image of the iris 11 can be acquired almost entirely, the amount of information is sufficient, but the eye 10 is not sufficiently large. In the case shown in (b) where it is not opened, the amount of information is insufficient, and it is impossible to identify the person himself / herself. You can ask your eyes to open wide for identity verification, but you want to get the data as natural as possible. Therefore, in the present invention, in addition to the iris data, the white eye data of the white eye portion extending to the left and right thereof is also acquired, so as to compensate for the lack of information.

FIG. 6 is an explanatory diagram of the structure of the eye. As shown in this figure, the eye has a pupil 17 at the center, and the iris 11 surrounds the pupil 17. In the iris 11, there are an iris crimp wheel 18 and a crypt 19, which distinguish each person's characteristics. Furthermore, a capillary 21 exists in the white part around the iris 11, and this is used as auxiliary information for personal identification this time.

FIG. 7 is a diagram for explaining the setting contents of the white-eye analysis band. As for the white-eye data, as shown in this figure, the analysis band 22 is set, and digital coding is performed in the same manner as when the iris data is just obtained.

Returning to FIG. 1, a method of using the iris data and white-eye data obtained as described above will be described. In the case of the present invention, the registration data for personal identification has, for example, a configuration as shown in FIG. That is, the registration data 1 includes the iris data 3, the white-eye data 5, and the registration date and time 7. The contents of the iris data 3 and the white-eye data 5 are as described above. The registration date and time 7 is the date and time when the white-eye data 5 was acquired.

Here, a method of personal identification using such registration data will be described. First, when performing personal identification, the similarity between registered data and actually acquired data is determined. As shown in FIG.
And In this case, the overall similarity D is calculated as shown in FIG. The iris similarity Da is obtained from the degree of coincidence of the iris data. The degree of coincidence is calculated from, for example, the ratio of the same bit to the entire digitally encoded data obtained under the same conditions. Further, the white-eye data similarity Ds is calculated by the same method. The white-eye data similarity Ds is multiplied by a coefficient α for weighting. Then, the sum of the iris similarity Da and the α × white-eye data similarity Ds is set as the overall similarity D.

The value of α is, for example, 0.
It is about 2 to 0.3. That is, by suppressing the weight of the white-eye data to about 20 to 30% lower than that of the iris data, it is possible to utilize the effect of performing personal identification mainly by using the iris data and using the white-eye data in an auxiliary manner.
The white-eye data has a stable personal identification function for a short period of time, but may change after a long period of time. Therefore, the registration date and time 7 is included in the registration data 1, and the elapsed time from the registration time is checked. And
If a certain period of time has passed since registration, re-registration will be performed. Further, the weight of the white-eye data similarity Ds is reduced as the time elapses from the time of registration, thereby preventing the personal identification result from being largely influenced by the change. Specifically, as shown in FIG. 1 (c), good results can be obtained by reducing α in proportion to the passage of time and re-registering about six months after the registration.

Hereinafter, the procedure of the process of photographing the eye to obtain iris data and white eye data and the process of identifying the person will be described in detail. FIG. 8 shows a flowchart of the eye photographing operation. As shown in this figure, first, the vicinity of the iris of the eye and the vicinity of the white eye are photographed. First, in step S1,
Eye shape recognition is performed. That is, when the person approaches, for example, a gate or the like, the operation of recognizing the position of the eyes of the person, recognizing the shape of the eyes, and focusing on the eyes is performed. At this time, first focus on the center of the eye in step S2,
In step S3, an infrared light source is irradiated. Such a light source is preferable because Japanese irises have a color from black to almost brown. In step S4, the iris is thus photographed. Further, in step S5, the boundary of the white eye is recognized, and in step S6, the boundary of the white eye is focused. And
In step S7, white light is emitted. When photographing the white blood capillaries, the capillaries are red, and the photographing becomes clearer when white light is irradiated. In this way, the focus is changed, and the white eye is photographed in step S8 using another light source.

FIG. 9 is a flowchart showing a processing operation using only iris data. This flowchart shows processing when sufficient iris data is obtained when an eye is actually photographed for personal identification. First, in step S1, eye image data is obtained, and in step S2, the shape of the eye is determined. The image data acquisition process in the first step S1 is as described with reference to FIG. Then, in step S2, the eye shape is determined. This is because the eyes are wide enough, and FIG. 5 (a)
This is to determine whether the information amount of the iris data is sufficient as shown in FIG. In step S3, it is determined whether or not the eyes are sufficiently open, that is, whether or not the eyes are narrow. If the eyes are sufficiently open, the process proceeds to step S4, and the iris portion is extracted. Then, in step S5, the pupil center is determined,
In step S6, the analysis band of the iris is determined. The processing of this side is the same as that described with reference to FIGS. 2, 3, and 4. In step S7, the shade of the iris is extracted.
At step 8, the iris is digitally encoded. In step S9, a registered pattern is input, and in step S10, the matching is performed. In step S11, the degree of similarity is determined. If not similar, the process proceeds to step S12, and it is determined that the user is not the person. If they are similar, the process proceeds to step S13, and the person is determined to be the person. On the other hand, if it is determined that the eyes are thin and the eyes are not sufficiently opened, the process proceeds from step S3 to the white-eye data use process.

FIG. 10 is a flowchart showing the operation of using white eye data. First, in step S1, an image of an iris and a white part is extracted, and in step S2, a pupil center is determined. In step S3, the analysis band of the iris and the white of the eye is determined, and in step S4, the shading of the iris is extracted. In step S5, the iris is digitally encoded,
In step S6, shading of white eyes is extracted. Step S7
Then, digital coding of the white eye is performed, thereby obtaining data for personal identification including the white eye portion. Next, in step S8, the registration data shown in FIG. 1 is input, and in step S9, the matching is performed. As described with reference to FIG. 1, the matching method weights and adds the iris similarity and the white-eye data similarity. As a result, in step S10, when it is determined that they are similar, the process proceeds to step S12, and the user is determined to be the same. If it is determined that they are not similar, step S11
Then, it is determined that the user is not himself.

<Effects of Individual Identification Method> As described above, when the eyes are not sufficiently open and the amount of information for identifying an individual using only the iris data is insufficient, the white-eye data is deleted. It can be used supplementarily for identification.

<Personal Identification Device> FIG. 11 is a block diagram showing a specific example of a personal identification device for implementing the method of the present invention. This apparatus includes a camera 31, an iris data acquisition unit 32, a white-eye data acquisition unit 33, a clock unit 34, a registration date and time acquisition unit 35, a registration data storage unit 36, a collation unit 37, a determination result display unit 38, and an elapsed time calculation unit 39. And a re-registration request unit 40. The camera 31 is for photographing an image or the like of a person entering the room. Iris data acquisition unit 32
Is a part for obtaining iris data by the method described above, and the white eye data obtaining unit 33 is a part for obtaining white eye data in the same manner. The clock unit 34 includes a digital counter and the like for displaying the current time in order to determine the registration date and time.

The registration date and time acquisition unit 35 is a device that includes the date and time when the white eye data was acquired with reference to the clock unit 34 in the registration data. The registration data storage unit 36 stores the registration data as shown in FIG. The collation unit 37 receives the iris data and white-eye data of the person who has actually entered the room from the iris data acquisition unit 32 and the white-eye data acquisition unit 33, collates the registered data with the registered data stored in the registered data storage unit 36, and performs similar processing. This is the part that makes the decision. The result is displayed by the determination result display unit 38. The determination result display section 38 is composed of a display and other display devices. The elapsed time calculation unit 39 reads the registration date and time 7 from the data read from the registration data storage unit 36 and compares it with the current time.
Notify 0 to that effect. The re-registration request unit 40 is a part that performs control to output the fact to a display, a printer, or the like in order to prompt the re-registration of the white-eye data with respect to the registered data that has passed a predetermined period or more from the registration. With the apparatus having the above configuration, the method of the present invention as described above can be performed.

<Effects of Personal Identification Device> By adding such a device to a device that performs entry management, transaction management, and the like, it is possible to more reliably compare the device with the device that uses only iris data.
More individual identification becomes possible. Further, it is possible to prevent the burden on the staff from increasing due to the result such as the inability to identify the individual.

[Brief description of the drawings]

FIG. 1 is an explanatory diagram of a personal identification method of the present invention.

FIG. 2 is an explanatory diagram of a method of determining a pupil center.

FIG. 3 is an explanatory diagram of setting contents of an analysis band of an iris.

FIG. 4 is an explanatory diagram of an example of digital coding of light and shade.

FIG. 5 is a diagram illustrating the relationship between an eye image and iris data.

FIG. 6 is an explanatory diagram of an eye structure.

FIG. 7 is an explanatory diagram of setting contents of a white eye analysis band.

FIG. 8 is a flowchart of an eye photographing operation.

FIG. 9 is a flowchart of a processing operation of only iris data.

FIG. 10 is a flowchart of a white eye data use processing operation.

FIG. 11 is a block diagram of a personal identification device according to the present invention.

[Explanation of symbols]

 3 Iris data 5 White eye data 7 Registration date and time

Claims (8)

[Claims] 1. An iris data having extracted features of the iris and a white eye data having extracted features of a white eye around the iris are registered and registered in advance, and the iris data obtained for personal identification and the registered iris data are registered. , Calculating the similarity of the iris data, comparing the white eye data acquired for personal identification with the registered white eye data, calculating the similarity of the white eye data, and calculating the similarity of the iris data. A personal identification method characterized in that, when the degree of similarity between the degree and the white-eye data is at a certain level or higher, the iris data and the white-eye data acquired for personal identification are determined to be those of the individual. 2. The method according to claim 1, wherein the total similarity is obtained by adding the similarity of the iris data and the similarity of the white-eye data. The iris data and the white-eye data acquired are determined to be those of the person, and a weighting coefficient is applied so that the similarity of the iris data has a greater effect on the overall similarity than the similarity of the white-eye data. Personal identification method. 3. The personal identification method according to claim 2, wherein a weighting coefficient is selected such that a ratio affecting the overall similarity of the white-eye data gradually decreases with time. 4. The personal identification method according to claim 1, wherein when a predetermined time has elapsed, white eye data is newly collected and re-registered. 5. The method according to claim 1, wherein when the shape of the eye photographed for personal identification is large enough to acquire a sufficient amount of iris data, the similarity of only the iris data is used for personal identification. It is determined that the iris data acquired is the person's own, and if the shape of the eye is too large to acquire a sufficient amount of iris data, it is acquired for personal identification by the similarity between the iris data and the white-eye data. A personal identification method characterized in that the determined iris data and white eye data are determined to belong to the person. 6. A registered data storage unit for registering and storing iris data obtained by extracting features of the iris and white eye data obtained by extracting features of a white eye around the iris; An iris data acquisition unit that acquires iris data, a white eye data acquisition unit that acquires white eye data, and accepts the outputs of the iris data acquisition unit and the white eye data acquisition unit and compares them with the corresponding data in the registered data storage unit. Calculating the similarity between the iris data and the white-eye data, and when the similarity between the iris data and the white-eye data is at a certain level or more, the iris data and the white-eye data acquired for personal identification. A personal identification device comprising: 7. The registered data storage unit according to claim 6, wherein the registered data storage unit registers the iris data, the white-eye data, and the registration date and time of the white-eye data. Is added to obtain the overall similarity, and when the overall similarity is at a certain level or higher, the iris data and the white-eye data acquired for personal identification are determined to be those of the person, and the iris data Weighting coefficients are applied so that the similarity has a greater effect on the overall similarity than the similarity of the white-eye data, and
Personal identification characterized by comprising an elapsed time calculation unit for selecting a weighting coefficient so as to gradually reduce the rate of affecting the overall similarity of the white-eye data with the passage of time and calculating the time elapsed from the registration date and time. apparatus. 8. The personal identification device according to claim 6, wherein the camera uses different illuminations at different focal points when capturing an iris image and when capturing a white-eye image. apparatus.