JP2000137811A - Individual identifier - Google Patents

Abstract

PROBLEM TO BE SOLVED: To decrease the number of dictionaries for collation and to increase the processing speed with respect to an individual identifier. SOLUTION: A representative dictionary collation part 2 stores the dictionaries which represent the features of each of groups including the dictionaries having their degrees of difference close to each other. A dictionary storing part 5 stores the dictionaries belonging to each group of representative dictionaries. The part 2 collates an input image with all representative dictionaries and decides a representative dictionary having the smallest degree of difference as a group of the input image. A dictionary selection part 4 reads the dictionary of a group corresponding to the representative dictionary at the part 2 out of the part 5. A collation part 6 collates the input image with the dictionary that is read at the part 4 to decide the genuineness of the input image.

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to an individual identification apparatus for identifying individuals based on biological characteristics, and more particularly, to an individual identification device.
For example, the present invention relates to an animal individual identification device for registering data for individual identification of animals and performing collation in an organization requiring individual management of animals such as (race) horses and (pastoral) cows.

[0002]

2. Description of the Related Art Individual identification of thoroughbreds at a racetrack, auction field, or a breeding ranch is carried out under the initiative of a pedigree control agency. The current method of identifying individuals in Japan is based on coat color, vitiligo on the limbs, vitiligo on the head, and curl. In the world, there are other methods based on blood type, branding, tattooing, etc. Reference: "Knowledge of Horses", II-
3. How to identify a horse, P. 153-P. 154).

[0003] In the case of individual identification of cattle, a method using identification nameplates such as collars and ear tags and branding and tattooing has been adopted in general ranches.

[0004] As an individual identification method of an animal that has recently attracted attention, there is a technique using an MC (Micro Chip). This is a chip with a built-in microminiature integrated circuit enclosed in a glass tube, which is implanted in the body of an animal with an injector such as a syringe.In the case of identification, the embedded signal is traced by a non-contact detector to trace the output signal. This is used as information for individual identification.

[0005]

However, the above-mentioned prior art has the following problems to be solved.
In the case of discrimination based on coat color, white mark, vitiligo, and curl, there are many horses with few features and horses with the same feature, so that the identification of the horse is often incomplete. In addition, there is a risk that the branding and tattoos may disappear or be tampered with, and improvements in animal welfare such as pain and local suppuration received by animals have been desired. In addition, although individual identification based on blood type is accurate, there is a disadvantage that a considerable processing time is required for determination and the cost is high.

[0006] The identification nameplate has a risk of being damaged, lost or stolen, and in the case of branding or tattooing, it has a risk of disappearing or being tampered with like a horse. ing. In particular, even if this falsification is revealed, it is difficult to prove where the person originally belonged.

[0007] The MC method can be used semi-permanently once it is embedded, and also has advantages such as high convenience. However, the operability of implantation in living animals, pain given to animals at the time of implantation, local reactivity such as swelling, tenderness, suppuration, etc., motor dysfunction and clinical abnormalities in animals, movement of MC in vivo There are many issues in terms of performance, operability of detector, change in detection sensitivity, stability, and reliability. And above all, from the aspect of animal welfare, there are animal officials who are reluctant to adopt the MC method,
A convenient identification method that replaces the MC identification method has been desired.

[0008]

The present invention employs the following structure to solve the above-mentioned problems. <Structure 1> Representation dictionary storage unit for storing representative dictionaries indicating information representative of the characteristics of each group, and storing dictionaries for storing dictionaries belonging to each representative dictionary group. Representative dictionary information that matches the information of the individual to be identified with all the representative dictionaries in the representative dictionary storage unit and determines the representative dictionary having the smallest difference from the information of the individual to be identified as a group of the individual to be identified. And a dictionary selection unit that reads a dictionary of a group corresponding to the representative dictionary determined by the representative dictionary matching unit from the dictionary storage unit, and a matching unit that performs a comparison between the dictionary read by the dictionary selection unit and determines whether the dictionary is authentic. An individual identification device characterized by comprising:

<Structure 2> A dictionary storage unit storing a dictionary indicating image information corresponding to each ID, and having information indicating whether or not each dictionary is excluded at the time of matching for each dictionary; A pixel selection table indicating, for each selection order, a pixel to be selected when the image of the dictionary is compared with the image of the dictionary, a threshold value table indicating a threshold provided corresponding to the selection order, and a selection order based on the pixel selection table. A pixel selection unit that reads pixels according to the above from a dictionary that is not excluded from the dictionary storage unit, a pixel of each dictionary read by the pixel selection unit, and a pixel of the image of the identification target individual, and the degree of difference is determined by a threshold. If the dictionary is larger than the threshold value in the table, information indicating that it is to be excluded is added, and matching is performed according to the selection order. The image to be identified individual against the ID of the dictionary identification device being characterized in that a and determining dictionary limited part is true.

[0010]

DESCRIPTION OF THE PREFERRED EMBODIMENTS Embodiments of the present invention will be described below in detail with reference to specific examples. The invention relates to (race) horses,
(Maki) In order to identify animals at an organization that requires individual management of animals such as cattle, for example, register data on iris and iris granules and register data for collation processing.
The present invention relates to a verification device. Hereinafter, in each specific example, a case where a target animal is a horse in particular will be described as an example.

FIG. 2 shows a front view of the eyeball of a horse. As shown in FIG. 2, the horse's eye mainly includes a pupil 10, an iris (iris) 11, and an iris granule 12 between upper and lower eyelids 13.

A transparent hemispherically raised cornea is provided in front of the pupil 10 and the iris 11. The major difference between horse eyes and human eyes is that the pupil 10 is elliptical,
It is to have iris granules 12 unique to horses and ruminants.

Light from the outside passes through the pupil 10 and reaches the retina at the back of the pupil. The iris 11 is a muscle surrounding the pupil 10 and has a function of controlling the amount of light incident on the pupil 10 by contracting and expanding. The iris granules 12 are located between the iris 11 and the pupil 10 and have a shape in which hemispherical raised granules are connected. The iris granules 12 have a black color rich in melanin pigment, and are said to have a function of absorbing dazzling daylight even when the iris 11 contracts.

The size and shape of the iris 11 and the iris granules 12 or the fine wrinkles and irregularities on the surface differ depending on the individual of the horse and the right and left eyes, and there are individual differences. for that reason,
The iris 11 and the iris granules 12 are registered as reference data (hereinafter, referred to as a "dictionary") by imaging the horse's eyes with a camera, and the iris 11 and the iris granule data obtained at the time of the examination are compared and compared with the dictionary. It is possible to identify the target individual.

[0015] However, in a large-scale collation apparatus having several hundreds of registered horses, it takes a long time to search for a horse having the same ID as the input image from all the registered horses.

The present invention, on the other hand, is intended to reduce the number of matching dictionaries in the individual identification device and speed up the processing. Hereinafter, the individual identification device of the present invention will be described in detail using specific examples.

<< Specific Example 1 >> In specific example 1, the dictionaries are classified into several groups based on similarity in advance, and a group selection function is provided by collating each group with a representative dictionary, thereby reducing the number of collation dictionaries. The processing speed is increased.

<Configuration> FIG. 1 is a configuration diagram showing a specific example 1 of the individual identification device of the present invention. The specific example 1 includes an imaging unit 1, a representative dictionary matching unit 2, a representative dictionary storage unit 3, a dictionary selecting unit 4, a dictionary storage unit 5, a matching unit 6, and a display unit 7. The representative dictionary storage unit 3, the representative dictionary matching unit. It is characterized in that a section 2 and a dictionary selecting section 4 are provided.

The image pickup unit 1 comprises an image input unit such as a CCD video camera and an illumination device for certifying an individual.
It captures the image of the horse's eyes as a signal. The image input unit is a part that captures an image of a horse's eye with a camera, and performs a process of converting analog optical information of the image of the horse's eye into a digital electric signal by a sensor. Also, the lighting device is provided to illuminate the horse's eyes in order to capture it with good contrast,
As a light source, light having a wavelength that is hardly perceived by an animal, such as red light, near-infrared light, or infrared light, is used to prevent the horse from dazzling and moving excessively.

The representative dictionary matching unit 2 has a reading function of the representative dictionary storage unit 3 and a difference calculating function. That is, the representative dictionary matching unit 2 compares the information of the identification target individual with all the representative dictionaries in the representative dictionary storage unit 3 and determines the representative dictionary having the smallest difference from the information of the identification target individual. Has the function of determining the group.

The representative dictionary storage unit 3 is a list including group numbers and image data of representative points. In other words, the representative dictionary storage unit 3 is a database that stores the representative dictionaries indicating information representative of the characteristics of each group by grouping those having similar dictionary information values.

The dictionary selecting section 4 has a function of reading a dictionary of a group corresponding to the representative dictionary determined by the representative dictionary matching section 2 from the dictionary storing section 5.

The dictionary storage unit 5 is an image database of registered horses, and is a list including IDs and image data provided for each group number. That is, the dictionary storage unit 5
This is a database that stores dictionaries belonging to each group of the representative dictionaries.

The collating unit 6 has a function of selecting dictionary data of a specific group from the dictionary storing unit 5 and reading the dictionary data into a memory (not shown), and a collating function of comparing the horse data obtained by the imaging unit 1 to calculate the degree of difference. , And a true / false judgment function for judging whether or not the horse is the same as the dictionary based on whether the difference is within the threshold value.

The display unit 7 has a function of displaying the result of determination of the authenticity and the degree of difference received from the output result of the collation unit 6, and is composed of an output device such as a display or a printer.

In the above specific example, the representative dictionary matching section 2, dictionary selecting section 4 and matching section 6 are realized by a processor executing programs corresponding to the respective functions. Further, such a program is provided by being recorded on a recording medium.

<Operation> FIG. 3 is a flowchart showing the operation of the first embodiment of the present invention. The steps will be described below in order.

[Step S1] The operation of step S1 is performed by the imaging unit 1. The horse's eyes are illuminated by an illuminator using red, near-infrared, or infrared light as a light source, and imaged by an image input device such as a CCD video camera. The CCD sensor is configured by a two-dimensional array of photoelectric conversion elements, and exposes reflected light of an imaging target for a certain period of time, and quantizes a voltage level imprinted by the photoelectric conversion at a certain gradation to convert it into a digital signal. Thus, a two-dimensional array of image frames is obtained.

[Step S2] Steps S2 and S3 are performed by the representative dictionary matching unit 2. The representative dictionary matching unit 2 first reads the representative dictionaries of all the groups from the representative dictionary storage unit 3. The representative dictionary is a dictionary storage unit 5
The images having the IDs registered in the... Group having similarities are grouped together, and the images representative of the characteristics of the group are selected. Various methods such as clustering can be applied to this specific example for forming a group at the time of dictionary registration.
The representative dictionary need not be the image itself, but may be a characteristic pixel selected for speeding up the representative dictionary comparison.

[Step S3] Next, the representative dictionary matching unit 2
Calculates the difference between the input image and all the representative dictionaries. The calculation of the degree of difference here is for classifying the input image into one of the groups, not for the authenticity determination as performed in steps S5 and S6 later.

[Step S4] Step S4 is performed by the dictionary selecting section 4. The dictionary selection unit 4 determines in step S
A group having the smallest degree of difference from each representative dictionary calculated in step 3 is selected as a group to which the same class as the input image belongs;
The dictionary of this group is read from the dictionary storage unit 5.

In this way, the number of matches with the dictionaries can be reduced from all dictionaries to the number of representative dictionaries + the number of dictionaries in a group. Therefore, this embodiment is particularly effective in an identification device having a large number of registrations. In addition, instead of selecting only the minimum group, a method of limiting to a plurality of groups equal to or less than a required threshold value may be used.

[Steps S5 and S6] Steps S5 and S6 are performed by the matching section 6. Collation unit 6
Performs true / false determination based on the degree of difference. The dissimilarity is compared with a predetermined threshold value. If there is a dictionary whose dissimilarity is equal to or less than the threshold value, it is determined that the inspection target is true; otherwise, it is false. If the result of the truth determination is “true”, the process proceeds to the dictionary registration process of step S7, and if the result is “false”, the process proceeds to the rejection process of step S8.

[Step S7] The collation unit 6 displays on the display unit 7 that the identification result is “accept”, and ends the processing.

[Step S8] The fact that the identification result is rejection is displayed on the display unit 7, and the processing is terminated.

<Effects> As described above, in the specific example 1, the dictionaries are classified into several groups in advance based on the similarity.
First, matching with the representative dictionary of each group is performed, and then matching with the dictionary of the specified group is performed. Therefore, the following effects can be expected.

In the animal individual identification device, the number of dictionaries to be collated can be reduced from all dictionaries to the number of representative dictionaries + the number of dictionaries in a group, so that the processing can be speeded up, and is particularly effective in an identification device with a large number of registrations. It is.

<< Specific Example 2 >> The individual identification device described in the specific example 1 aims to speed up the processing by reducing the number of dictionaries to be collated. On the other hand, the specific example 2 is an individual identification device in which the collation time with each dictionary is reduced. More specifically, when calculating the degree of difference, matching is performed preferentially from the characteristic pixels of each dictionary, and during the accumulation of the degree of difference of each pixel, the matching of a particularly large degree of difference is interrupted, so that the matching process is performed. It is intended to increase the speed.

<Structure> FIG. 4 is a diagram showing the structure of a specific example 2 of the individual identification apparatus according to the present invention. Specific example 1 is an imaging unit 10
1, pixel selection unit 102, pixel selection table 103, dictionary storage unit 104, dissimilarity accumulation unit 105, dictionary limitation unit 10
6, comprising a threshold table 107 and a display unit 108, wherein a pixel selection table 103 and a pixel selection unit 102, and a threshold table 107 and a dictionary limiting unit 106 are provided.

As in the first embodiment, the imaging unit 101
It comprises an image input unit such as a D video camera and an illumination device for certifying an individual, and captures the image of a horse's eye as a signal. The image input unit is a part that captures an image of a horse's eye with a camera, and performs a process of converting analog optical information of the image of the horse's eye into a digital electric signal by a sensor. In addition, the lighting device is provided to illuminate the horse's eyes in order to catch the eye with good contrast. As a light source, in order to prevent the horse from dazzling and excessively moving, it emits red light, near infrared light, or infrared light. In such animals, light having a wavelength that is difficult to detect is used.

The pixel selection section 102 has a pixel selection table 1
A function of reading a pixel number from 03 into a memory (not shown),
And a function of reading the pixel value of the corresponding pixel number from the dictionary storage unit 104. Here, the dictionaries to be read from the dictionary storage unit 104 are only those for which the matching is not excluded.

The pixel selection table 103 is a list of pixel selection order and pixel numbers. That is, the pixel selection table 103 is a table indicating, for each selection order, the pixels to be selected by the dictionary limiting unit 106 when matching the image of the individual to be identified with the image of the dictionary. Here, the relationship between the selection order and the pixel number is determined, for example, as follows.

FIG. 5 is an explanatory diagram when the pixel selection table 103 is created.

For example, a characteristic value for an arbitrary pixel in a certain dictionary can be represented by “average difference from other eyes / average difference between same eyes”. Here, the average degree of difference from other eyes is the following value.

FIG. 5A shows an image of each dictionary stored in the dictionary storage unit 104, and each grid portion represents a pixel. It is assumed that dictionaries A, B,..., Z are prepared.

For example, when attention is paid to the pixel number i, the difference between the pixel value of the pixel number i of the dictionary A and the pixel value of the pixel number i of the dictionary B, the pixel value of the pixel number i of the dictionary A and the pixel value of the dictionary C The difference between the dictionary A and each of the dictionaries B to Z, such as the difference between the pixel values of the number i and the pixel value of the pixel number i of the dictionary A and the pixel value of the pixel number i of the dictionary Z, etc. Then, the average degree of difference is obtained.

Further, the average difference of each pixel in the dictionary A is obtained from a plurality of images relating to the same dictionary (for example, the dictionary A). That is, the average difference of the pixel values (the difference between the image A1 and the image A2, the difference between the image A1 and the image A3,...) Is obtained from the plurality of images for the pixel number i. Note that the plurality of images related to the same-eye dictionary may be, for example, a plurality of images used when registering a dictionary.

As a result, it is possible to obtain a feature value relating to the pixel number i in a certain dictionary. FIG. 5 shows the characteristic values of the pixel numbers for each dictionary thus obtained.
(B). Here, the characteristic values of the dictionary A and the dictionary B are shown.

FIG. 5C shows the pixel selection table 103. In the pixel selection table 103, the selection order is set in the order of the pixel number having the largest feature value for each dictionary. For example,
In the case of the dictionary A, the pixel number 3 has the largest feature value, and the pixel number 1 is the next. Therefore, the selection order is set to this order.

In this example, to simplify the description, the order of selection in the pixel selection table 103 is common to each dictionary.

The dictionary storage unit 104 is a registered horse database including IDs, image lists, and flags. That is, the dictionary storage unit 104 is a database that stores a dictionary indicating image information corresponding to each ID, and has a flag indicating whether or not each dictionary is to be excluded at the time of matching, for each dictionary. In this specific example, a setting is made such that when the flag is 1, it is excluded at the time of collation.

The dissimilarity accumulating unit 105 has a function of calculating the dissimilarity between the input image and the pixel value of each dictionary in the dictionary storage unit 104, and performing cumulative addition.

The dictionary limiting unit 106 includes a threshold table 107
A reading function of a threshold value corresponding to a selection order, a comparison function of a cumulative addition value of a degree of difference and a threshold value, a dictionary selection function,
It consists of a true / false judgment function. That is, the dictionary limiting unit 106 checks the pixels of each dictionary read by the pixel selecting unit 102 with the pixels of the input image, and excludes a dictionary whose difference degree is larger than the threshold value of the threshold value table 107. Is given according to the selection order, and if there is a dictionary with a dissimilarity smaller than the final order threshold value in the comparison with the final order pixel, the input image is true for the dictionary ID. It has a function to determine that there is.

The threshold table 107 is a list including a selection order and a threshold. FIG. 6 is an explanatory diagram showing the relationship between the selection order of the threshold value table 107 and the threshold value. In this specific example, the threshold value is large in proportion to the selection order because the difference is compared with the value accumulated for each selection order.
Also, at the stage where the selection order is small (the number of collation is small),
In order to prevent the dictionary having the same ID from being excluded from the target, the threshold value is set to be looser than the condition where the threshold value is simply divided by the number of pixels. When the selection order is the final order N, the threshold value is the authenticity determination threshold value.

The display unit 108 is an output device such as a display or a printer having a function of displaying the result of determination of the authenticity and the degree of difference received from the result of the dictionary limiting unit 106.

<Operation> FIG. 7 is a flowchart showing the operation of the second embodiment. The steps will be described below in order.

[Step S101] Step S101 is performed by the imaging unit 101. The horse's eyes are illuminated by illumination using red, near-infrared, or infrared light as a light source, and are imaged by an image input device such as a CCD video camera. The CCD sensor is constituted by a two-dimensionally arranged photoelectric conversion element, and exposes the reflected light of the imaging target for one transit time, and converts the voltage level imprinted by the photoelectric conversion into a digital signal by quantizing it at a constant gradation. Thus, a two-dimensional array of image frames is obtained.

[Step S102] Step S102,
S103 is performed by the pixel selection unit 102. The pixel selection unit 102 first reads the pixel number of the corresponding selection order from the pixel selection table 103. Selection order is initial value 1
, And is incremented by 1 in a later step S107.

[Step S103] Pixel selection unit 102
Are all IDs from the dictionary storage unit 104 for this pixel number.
Read the pixel values of the dictionary. At this time, the dictionary whose flag is 1 is excluded from the dictionary because it is a dictionary excluded from the collation in step S106.

[Step S104] Step S104 is performed by the dissimilarity accumulating section 105. Difference degree accumulator 1
In step 05, the degree of difference between the input image and the dictionary of each ID is determined for the selected pixel, and the difference is cumulatively added.

[Step S105] Steps S105 to S105
S107 is performed by the dictionary limiting unit 106. The dictionary limiting unit 106 reads the threshold of the corresponding selection order from the threshold table 107.

[Step S106] Next, the dictionary limiting unit 1
06 compares the cumulative value of the dissimilarity with the threshold value, and, for a dictionary whose cumulative value is greater than the threshold value,
The flag in the dictionary of D is set to 1 and excluded from the comparison target.
Note that the threshold is set to a modest condition as shown in FIG. 6, and thus it is unlikely that a dictionary having the same ID as the input image is excluded.

[Step S107] If there is no dictionary in step S106 in which the accumulated value is equal to or smaller than the threshold, the dictionary limiting unit 106 determines that there is no dictionary corresponding to the input image and proceeds to the rejection process in step S109. When the selection order is the final order, the threshold value is the authenticity determination threshold value as described above. Therefore, if there is a dictionary in which the cumulative value is equal to or smaller than the threshold in the final rank, the process proceeds to step S108 for accepting the dictionary. Otherwise, the selection order is set to 1 for the next pixel selection.
It increases and it progresses to step S102.

[Step S108] The display unit 108 displays that the identification result is "accept", and ends the processing.

[Step S109] The fact that the identification result is rejection is displayed on the display unit 108, and the processing is terminated.

<Effects> As described above, according to the specific example 2, when calculating the degree of difference, the matching is performed preferentially from the characteristic pixels of each dictionary, and the difference is calculated particularly during the accumulation of the degree of difference of each pixel. Since the collation of large items is interrupted, the following effects can be expected.

In the animal identification device, different I
Since the collation with the dictionary of D can be interrupted at an early stage, the effect of speeding up the collation processing can be expected.

<Usage Form> In the above description of the specific examples, the collation unit and the dictionary have been described as using images. However, the principle of the present invention is not limited to collation using various features extracted from images. The same applies.

In the description of the operation of the specific example 2, the selection order of the pixel selection table 103 is common to each dictionary.
The same applies to the case where the selection order is set for each dictionary as shown in FIG. Further, the relationship between the selection order and the pixel number may be set other than the description of the specific example 2.

Although the case of a horse as an individual to be identified has been described, the present invention is not limited to this, and other animals or humans can be similarly applied.

[Brief description of the drawings]

FIG. 1 is a configuration diagram showing a specific example 1 of an individual identification device of the present invention.

FIG. 2 is an explanatory diagram of an eyeball image of a horse.

FIG. 3 is a flowchart showing an operation of a specific example 1 of the individual identification device of the present invention.

FIG. 4 is a configuration diagram showing a specific example 2 of the individual identification device of the present invention.

FIG. 5 is an explanatory diagram when a pixel selection table is created in a specific example 2 of the individual identification apparatus of the present invention.

FIG. 6 is an explanatory diagram of a threshold table in a specific example 2 of the individual identification device of the present invention.

FIG. 7 is a flowchart showing an operation of a specific example 2 of the individual identification apparatus of the present invention.

[Explanation of symbols]

 2 Representative dictionary matching unit 3 Representative dictionary storage unit 4 Dictionary selection unit 5, 104 Dictionary storage unit 6 Matching unit 102 Pixel selection unit 103 Pixel selection table 105 Dissimilarity accumulation unit 106 Dictionary limitation unit 107 Threshold table

Claims (2)

[Claims] 1. A representative dictionary storage unit for storing representative dictionaries indicating information representative of characteristics of each group by grouping those having similar values of dictionary information, and a dictionary belonging to each representative dictionary group. The dictionary storage unit, the information of the identification target individual, and all the representative dictionaries of the representative dictionary storage unit are compared, and the representative dictionary having the smallest difference from the information of the identification target individual is identified as the group of the identification target individual. A representative dictionary matching unit to be determined; a dictionary selecting unit that reads a dictionary of a group corresponding to the representative dictionary determined by the representative dictionary matching unit from the dictionary storage unit; and a dictionary that is read by the dictionary selecting unit. And a collating unit for performing authenticity discrimination. 2. A dictionary storage unit for storing a dictionary indicating image information corresponding to each ID, and having information indicating whether or not each dictionary is excluded at the time of matching for each dictionary; Based on the pixel selection table, a pixel selection table indicating a pixel to be selected when the image is compared with the image of the dictionary for each selection order, a threshold value table indicating a threshold value provided corresponding to the selection order, A pixel selection unit that reads a pixel according to a selection order from a dictionary that is not excluded from the dictionary storage unit, and compares a pixel of each dictionary read by the pixel selection unit with a pixel of the image of the identification target individual, For a dictionary whose dissimilarity is larger than the threshold value of the threshold table, information indicating that the dictionary is excluded is added, and matching is performed in accordance with the selection order. If there is again dissimilarity dictionary, the identification target individual image identification apparatus characterized by comprising a determining dictionary limited portion to be true with respect to the ID of the dictionary.