JPH1125270A - Individual identifying device - Google Patents

Abstract

PROBLEM TO BE SOLVED: To obtain the designation of a horse by identifying each horse. SOLUTION: A horse to be identified is photographed from different directions by a picture inputting part 10, and when the picture of the horse is inputted, the picture analysis of the picture of the horse is operated by a picture analyzing means 20, and the appearance features of the horse such as the hair color of the horse or the pattern of the white dot of the head are obtained with each feature name. The designations of the plural horses and the feature names indicating the features are preliminarily stored so as to correspond to each other in a data base 31. A horse name outputting part 30 retrieves the data base 31 with the feature name applied from the picture analyzing means 20. The designation of the horse to be identified can be obtained by this retrieval.

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to an individual identification device for identifying an individual from an image of the individual.

[0002]

2. Description of the Related Art On a race day, a horse is inspected for a horse scheduled to run in a race. The following literature describes the horse body test. Reference: “All about Central Horse Racing”, first edition (S.40-12-25), published by Japan Racing Association, P.79,172-179. This is performed to execute a race, and performs individual identification of a racehorse, and checks for the presence or absence of a disease and gait. The horse body inspection is performed visually by a skilled inspector, and when the individual is identified, the horse is specified based on the coat color of the horse and other characteristics of the horse. The horse's coat color and other characteristics of the horse are defined by the Japan Light Stallion Registration Association Horse's coat color and characteristic description guidelines.

[0003]

However, in the conventional technique for identifying a racehorse, since the color and characteristics of the racehorse are judged by a person, the judgment result tends to be ambiguous, and the identification accuracy is uniform. It was difficult to keep. Also,
There is a problem that the possibility of erroneous determination such as oversight or the like cannot be denied because a person makes a determination.

[0004]

In order to solve the above-mentioned problems, a first aspect of the present invention is to provide an individual identification device, which captures an individual to be identified from different directions and obtains an image of the individual to be identified. An image input unit for inputting an image, image analysis means for analyzing the image to extract appearance characteristics of the individual to be identified, and an output unit for obtaining a name for identifying the individual to be identified based on the characteristic Are provided. According to a second aspect, in the individual identification device according to the first aspect, the image analysis means corresponds to the image from among feature names predefined and assigned to the variables for each element of the appearance. The feature is extracted by selecting one feature name for each element. The third invention is the second invention
In the individual identification device of the present invention, a database is stored in which data representing the appearance characteristics of a plurality of individuals using the feature names and the names representing the individuals are stored in association with each other, and the output unit includes: The database is searched, and the name of an individual having a feature name matching the feature name selected by the image analysis means is obtained as the name of the individual to be identified.

According to a fourth aspect, in the individual identification device according to the second or third aspect, the individual to be identified and the plurality of individuals are animals. And the image analysis means,
In the database and the output unit, one or more of the animal's coat color, head vitiligo pattern, curl position, and limb vitiligo pattern are used as the appearance elements. According to a fifth aspect, in the individual identification device according to the fourth aspect, the image analysis means extracts at least one of a position of a wound of the animal to be identified or a state of a spot other than vitiligo from the image. The database stores the names of the plurality of animals in association with the positions of the wounds or the states of the spots on the animals, and the output unit outputs the positions of the wounds or the states of the spots extracted from the image. Also, the database is searched by using the information processing apparatus to obtain the name of the animal to be identified. According to a sixth aspect of the present invention, in the individual identification device, name input means for inputting the name of the individual to be identified, and an image input unit for photographing the individual to be identified from different directions and inputting an image of the individual to be identified. Image analysis means for analyzing the image to extract the appearance characteristics of the individual to be identified, and accumulating the names of the plurality of individuals and data representing the characteristics of the appearance of each individual in association with each other, A database for searching and outputting the data corresponding to the name input through the name input means, and comparing the features searched for in the database with the features extracted by the image analysis means to determine the identification target. And an authenticity judgment unit for judging the authenticity of the individual.

According to a seventh aspect, in the individual identification apparatus according to the sixth aspect, the image analysis means selects one of the feature names defined and assigned in advance to the variable for each element of the appearance. The feature is extracted by selecting one feature name corresponding to the image for each element. In an eighth aspect based on the individual identification device according to the seventh aspect, the database associates data representing appearance characteristics of the plurality of individuals using the feature names with the names representing the individuals. The authenticity determination unit is configured to determine the authenticity of the individual to be identified by comparing the feature name searched in the database with the feature name selected by the image analysis unit. I have. The ninth invention is
In the individual identification device according to an eighth aspect, the individual to be identified and the plurality of individuals are animals. In the image analysis means, the database and the authenticity determination unit, one of the animal's coat color, head vitiligo pattern, curl position, or limb vitiligo pattern is used as the appearance element.
Use more than one. According to a tenth aspect, in the individual identification apparatus according to the ninth aspect, the image analysis means extracts at least one of a position of a wound of the animal to be identified or a state of a spot other than vitiligo from the image. The database accumulates the names of the plurality of animals in association with the positions of the wounds or the states of the spots on the animals, and the authenticity determination unit determines the positions or spots of the wounds extracted from the image. Is used to determine the authenticity of the animal to be identified.

According to an eleventh aspect, the first, second, third, fourth, fifth,
In the individual identification device of the invention of 6, 7, 8, 9 or 10, the image input unit is arranged around the individual to be identified, and is constituted by a plurality of cameras each of which captures the identification target. . The twelfth invention is directed to the first, second, third, fourth,
In the individual identification device according to any one of the fifth, sixth, seventh, eighth, ninth, and tenth aspects, the image input unit includes a single camera that moves around the individual to be identified and photographs the identification target. I have. According to the first to twelfth aspects, since the individual identification device is configured as described above, images obtained by photographing the individual to be identified from different directions are input from the image input unit. The image analysis of this image is performed by the image analysis means, and the characteristics of the horse to be identified are extracted. A horse to be identified is identified from the extracted features. Therefore, the above problem can be solved.

[0008]

DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS First Embodiment FIG. 1 is a block diagram showing the configuration of an individual identification apparatus according to a first embodiment of the present invention. This individual identification device is, for example, a device for identifying a racehorse that is an individual, and includes an image input unit 10.
An image analysis means 20 for analyzing an image input from the image input unit 10 to extract characteristics of a horse to be recognized; a horse name output unit 30 connected to the image analysis unit 20; The unit 20 includes a database 31 for providing characteristics of individual individuals. The image input unit 10 inputs an image of a horse to be recognized, and includes one or a plurality of cameras. The image analysis means 20 includes a CPU, a memory, and the like, and has a coat color identification section 21 that identifies a coat color of a horse from an image input from the image input section 10. On the output side of the hair color identification unit 21, a head vitiligo identification unit 22 is connected. A curl discrimination unit 23 is connected to the output side of the head vitiligo discrimination unit 22, and a limb vitiligo discrimination unit 24 is connected to the output side of the curl discrimination unit 23. Image analysis means 20
The horse name output unit 30 is connected to the output side of the limb vitiligo identification unit 24. The horse name output unit 30 outputs the identification result of the horse to be identified.

The operation of this individual identification apparatus is shown in FIGS.
This will be described with reference to FIG. FIGS. 2A and 2B are diagrams of FIG.
3A and 3B are diagrams illustrating an example of a camera position of the image input unit 10, wherein FIG. 3A is a front view and FIG. 3B is a plan view. 3A to 3C show the cameras 11 to 13 in FIG.
It is a figure which shows each horse image | photographed in. Image input unit 10
The cameras 11 to 13 are disposed at various places around the horse H as shown in FIGS. 2A and 2B in order to photograph the appearance of the horse H to be identified. Various optical parameters such as the focal length and the aperture of each of the cameras 11 to 13 are prepared according to the distance to the horse H of the subject and the magnification. As shown in FIG. 3A, the side camera 11 that captures the horse H from the side captures an image of the horse H's torso and leg sides. The front camera 12 arranged in front of the horse H, as shown in FIG.
The face, chest and front leg are photographed. The rear camera 13 arranged on the rear side of the horse H captures the buttocks and the rear of the legs of the horse H as shown in FIG. Note that FIGS. 2A and 2B illustrate an example in which a plurality of cameras 11 to 13 are used. However, one camera is moved and FIGS.
May be taken. The image input unit 10
The image of the horse to be recognized is output to the image analysis means 20.

FIG. 4 is a flowchart showing the processing of the coat color identification unit 21 in FIG. 1, and FIG. 5 is a diagram showing positions where the coat color is evaluated. FIG. 6 is a view showing feature names of horse hair colors. The coat color identification unit 21 refers to the color data 21-1 and coat color data 21-2 stored in a memory (not shown) and performs image analysis on the image input from the image input unit 10 in the processes S1 to S8 in FIG. Then, a feature name defined for the hair color, which is an appearance element of the horse, according to the variation is extracted and output as a hair color identification result. As shown in FIG. 6, the characteristic names of the coat color of the horse are defined as brown hair, tochi chestnut hair, deer hair, black deer, blue deer, blue deer, blue deer, and ash. The coat color identification processing S1 is performed from the given image.
The color information of the torso region 41 in FIG. 5 is extracted and compared with the color data 21-1 stored in advance. By this comparison, the color closest to the area 41 is determined, and this is extracted as the result of identifying the color of the fur. The color data 21-1 stores yellow-brown, red-brown, black, white-white, brown and the like as horse color data. As a method of expressing the color data 21-1, for example, an RGB color system composed of red (R), green (G), and blue (B) components in a pixel or CIE-X
A display method such as a YZ color system can be used.

In the long coat color identification processing S2 after the coat color identification processing S1, the color information of the mane area 42 and the tail coat area 43 in FIG. 5 is extracted from the given image, and the coat color identification processing S1 is performed. Similarly, the color of the long hair (mane and tail hair) of the horse H is extracted as the identification result. In the lower limb coat color identification processing S3, the color information of the leg areas 44 and 45 in FIG. 5 is extracted from the given image, and the coat color of the horse H legs is identified in the same manner as the coat color identification processing S1. Extract as a result. In the eye periphery hair color identification processing S4, the color information of the area 46 around the eyes in FIG. 5 is extracted from the given image, and the hair color around the eyes of the horse H is identified in the same manner as the coat color identification processing S1. Extract as a result. In the armpit hair color identification processing S5, the color information of the armpit area 47 in FIG. 5 is extracted from the given image, and the armpit hair color of the horse H is extracted as the identification result in the same manner as the coat color identification processing S1. Lower belly hair color identification processing S
In 6, the color information of the lower abdomen region 48 in FIG. 5 is extracted from the given image, and the lower abdomen hair color of the horse H is extracted as the identification result in the same manner as the coat color identification processing S1. In the nose periphery hair color identification processing S7, the color information of the area 49 around the nose in FIG. 5 is extracted from the given image, and the hair color around the nose of the horse H is identified in the same manner as the coat color identification processing S1. Extract as a result. In the coat color determination processing S8, the color identified in the processing S1 to S8 is compared with a combination of the colors stored in the coat color data 21-2, and the coat color of the horse H to be identified is determined. By this determination, it can be seen that the color of the hair of the horse H is “brown hair”, “brown horse chestnut”, “brown”, “black bristle”, “blue bristle”, “blue bristle” or “ash bristle”.

FIG. 7 is a flowchart showing the processing of the white spots identification section 22 in FIG. 1, and FIG. 8 is a diagram showing the extraction area of FIG. FIG. 9 is a diagram illustrating a horse head vitiligo pattern (No. 1), and FIG. 10 is a diagram illustrating a horse head vitiligo pattern (No. 2). FIG. 11 is a diagram showing feature names of head vitiligo. Referring to FIGS. 7 to 11,
The processing of the head white spot identifying unit 22 will be described. The head vitiligo identification unit 22 refers to the head vitiligo data 22-1 stored in a memory (not shown), and generates a white image for the image captured by the front camera 12 among the images provided from the image input unit 10. The region is extracted, image analysis is performed in steps S11 to S15 in FIG. 7, and a feature name defined according to a variable of the pattern of the white spot on the head, which is an appearance element of the horse, is obtained. Output as In the head vitiligo data 22-1, the feature names of vitiligo are accumulated as shown in FIG.

In the forehead white spot identification process S11 in FIG. 7, the white area of the forehead part area 51 in FIG. 8 is extracted from the given image, and the white area is referred to with the head white spot data 22-1. The feature name is selected by extracting the shape, size, and presence / absence of a void. According to the state of the vitiligo pattern of the forehead of the horse H to be identified,
The features are divided into "stars", "great stars", "song stars", "meteors" and the like. For example, when it is determined that the shape of the pattern is round, large in size, the number is 1, and there is no hollow, “great star” shown in FIG. 9 is selected as the feature name. If is small, “star” or “small star” is selected. Also,
If the shape is not a circle but has a flow below, “large meteor” in FIG. 9 is selected, and if the shape is curved, “song star” is selected as a feature name. If there is a hollow, "ring star" in Fig. 9
Is selected. In the nose bridge white spot identification processing S12, the white area of the nose bridge part 52 in FIG. 8 is extracted from the given image, the width of the white area is measured, and the head white spot data 22-1 is obtained.
And select the feature name in FIG. In the nose vitiligo identification process S13, a white area of the nose area 53 in FIG. 8 is extracted from the given image, and the size of the white area is measured.
The feature name in FIG. 10 is selected with reference to the head vitiligo data 22-1. In the lip vitiligo identification processing S14, the white area of the lip area 54 in FIG. 8 is extracted from the given image, the width of the white area is measured, and the white area of the lip is measured with reference to the head vitiligo data 22-1. 10. Select ten feature names. In the forehead white spot identification processing S15, the white area of the area 55 of the forehead on the forehead in FIG. 8 is extracted from the given image, the size and width of the white area are measured, and the head white spot data 22- 1 is selected with reference to FIG.

FIG. 12 is a flow chart showing the processing of the curl identification unit 23 in FIG. 1, and FIGS. 13 (a) to 13 (d).
FIG. 4 is a diagram illustrating positions where a triangular pattern is detected. FIG. 14 is a diagram showing the curl position and the curl name of the horse.
The curl identification unit 23 refers to the curl pattern data 23-1 and the curl position data 23-2 stored in a memory (not shown), performs the processes S21 and S22 in FIG. The characteristics of the curls of each horse are determined from the captured images. The curl pattern data 23-1 stores the curl pattern corresponding to the coat color of the horse, and the curl position data 23-2 accumulates the curl position as shown in FIG. ing. The number of the curl position in FIG. 14 is the number 6 assigned to each part of the horse H in FIG.
1 to 79. 12. The curl extraction process S21 in FIG.
Then, the spiral pattern corresponding to the coat color obtained by the coat color identification unit 21 is selected from the spiral pattern data 23-1, and the positions 61 to 79 where the spiral pattern exists are extracted from a given image. In the curl position identification process S22, a curl name is obtained from the curl position data 23-2 according to the curl positions 61 to 79 obtained in the curl extraction process S21.

FIG. 15 is a flowchart showing the processing of the limb vitiligo identification unit 24 in FIG. 1, and FIG. 16 is a diagram showing the extraction region of FIG. FIG. 17 is a diagram showing feature names of limb vitiligo. The limb vitiligo identification unit 24 refers to the limb vitiligo data 24-1 stored in a memory (not shown), extracts a white region from the image provided from the image input unit 10, and performs processing S31 in FIG. In S32, image analysis is performed to obtain a feature name defined in accordance with the variable regarding the pattern of the white spot on the limb, which is the appearance element of the horse, and output the result as a white spot identification result. In the limb vitiligo data 24-1, the feature names of the vitiligo are accumulated as shown in FIG. 17 according to the position and the feature amount of the vitiligo. Instep S31 of FIG.
In FIG. 16, the limb vitiligo identification unit 24 extracts a white area in the area of the footcap 81 at the base of the foot in FIG. 16, measures the size and the perimeter of the white area, and corresponds to the limb vitiligo data 24-1. Select the feature name to be used. For example, if the white area representing a vitiligo is small, a feature name of “slight white” is selected. In the lower extremity vitiligo identification processing S32, the extremity vitiligo identification unit 24 extracts a white area in the area 82 under the lower extremity in FIG. 16, measures the size of the white area, and measures the perimeter thereof. Select the corresponding feature name from.

FIG. 18 is a diagram showing horse names and feature names stored in the database 31 in FIG. The horse name output unit 30 searches the database 31 with each feature name selected by the image analysis unit 20. Since the names of a plurality of horses and the feature names of the respective horses are stored in the database 31 in association with each other, the database 31 is searched with the feature names selected by the image analysis means 20, and thereby the identification target is identified. You can ask for the name of the horse. For example, as shown in FIG. 18, the feature names selected and extracted in the processes of the respective units 21 to 24 of the image analysis unit 20 are “brown hair”, “meteor nose bridge white”, “shumoku”, and “right front small white”. In this case, the name of the horse is required as "Akasatana".

As described above, the horse H as the individual to be identified is
And an image input unit 10 for inputting an image of the horse H, and extracting features of the horse-specific appearance elements such as the coat color, head vitiligo, trichomes, limb vitiligo, etc. of the horse H from the image. , And the horse name output unit 30 that searches the database 31 and obtains the name of the horse H, the individual identification device is configured. Can be prevented, and stable horse identification can be performed. In addition, identification is possible without relying on skilled personnel. Further, since each feature is configured to be represented by a feature name and compared, for example, compared to a case where each horse is quantitatively represented, quantitative measurement of each horse is unnecessary, and the characteristics of each horse are easily stored in the database 31. Can accumulate. Then, the capacity of the database 31 can be significantly smaller than when storing images of a plurality of horses in the database 31.

Second Embodiment FIG. 19 is a block diagram showing the configuration of an individual identification apparatus according to a second embodiment of the present invention. This individual identification device includes an image input unit 10, an image analysis unit 20, and a database 31 similar to those of the first embodiment, and is a name input unit that does not exist in the first embodiment and that inputs the name of the horse H. A certain horse name input unit 90 and a true / false judgment unit 91 are provided. The authenticity judgment unit 91 is provided in place of the horse name output unit 30 of the first embodiment, and is connected to the output side of the image analysis unit 20 and the output side of the database 31. The horse name input unit 90 is configured to give the input horse name to the database 31. In the first embodiment, when the horse name of the horse H to be identified is unknown, the individual identification device obtains the name of the horse by extracting features from the photographed image. The individual identification device according to the embodiment is an individual identification device that determines whether the horse being imaged corresponds to the identification target horse H when the name of the identification target horse H is known in advance.

Next, a description will be given of a process for determining whether or not the horse being photographed corresponds to the horse H to be identified with the individual identification apparatus. First, the name of the horse to be identified is input to the horse name input unit 90. The horse name input unit 90 sends the input name to the database 31, and searches and selects a horse feature name corresponding to the name. On the other hand, the image input unit 10 and the image analysis unit 20 select the feature name of each element of the appearance of the horse H and extract the feature by the same processing as in the first embodiment. The authenticity determination unit 91 determines the feature name of the horse H retrieved from the database 31 and the image analysis unit 20.
Are compared, and if they match or approximate to each other, the identification target horse H being photographed corresponds to the horse whose name has been input to the horse name input unit 90. Output If they do not match, the horse name input section 9
It outputs that it does not correspond to the horse whose name is input to 0. As described above, in the individual identification device according to the second embodiment, the individual name identification device including the image input unit 10 and the image analysis unit 20 is provided with the horse name input unit 90 and the authenticity determination unit 91. When the name of the horse is known, the comparison between the feature obtained from the image and the feature obtained from the name is performed only once, so that the authenticity of the horse to be identified can be determined. Therefore, the time required to search for data when identifying a horse can be significantly reduced.

Note that the present invention is not limited to the above embodiment, and various modifications are possible. For example, there are the following modifications. (1) In the image analysis means 20 of the first and second embodiments, the hair color, the pattern of the vitiligo of the head, the position of the trichome, or the vitiligo pattern of the limb, which represents the characteristics of the horse H by the characteristic name, are extracted from the database 31. Are stored with the feature names corresponding to these. However, a flaw other than a feature represented by the feature name and a spot other than a white spot may be extracted by image analysis and searched in the database 31. (2) In the first embodiment, the database 31 is configured to obtain the name of the horse from the appearance characteristics of the horse. However, in addition to the appearance, the registration number, date of birth, blood type, pedigree registration, breed (Eg, thoroughbred male), producer name,
It is also possible to add a producer address, a production place, a father horse name, a mother horse name, and the like. (3) In the first and second embodiments, the individual identification device that identifies the horse H has been described. However, any device that has the same appearance characteristics as a horse can be applied to other animals. Needless to say, if you change the elements and feature names of the appearance features,
It can be applied to other than animals.

[0021]

As described in detail above, the first to fifth embodiments
According to the invention, an image input unit that captures an image of an individual to be identified from different directions and inputs an image, and an image that performs image analysis on the image input from the image input unit and extracts features of the individual to be identified Since the analysis means and the database constitute the individual identification device at the output unit, the identification of the individual such as a horse can be performed stably without overlooking. 6th to 1st
According to the invention, the image input unit, the image analysis unit, the name input unit, and the authenticity determination unit are provided in the individual identification device. Therefore, when the name of the individual is known, the identification target individual can be easily identified. True / false can be determined.

[Brief description of the drawings]

FIG. 1 is a configuration block diagram of an individual identification device according to a first embodiment of the present invention.

FIG. 2 is a diagram illustrating an example of a camera position of the image input unit 10 of FIG.

FIG. 3 is a view showing a horse photographed by cameras 11 to 13 in FIG. 2;

FIG. 4 is a flowchart showing a process of a coat color identification unit 21 in FIG. 1;

FIG. 5 is a diagram showing positions where hair color is evaluated.

FIG. 6 is a diagram showing feature names of horse hair colors.

FIG. 7 is a flowchart showing processing of a white spot identification section 22 in FIG. 1;

FIG. 8 is a diagram showing an extraction region of FIG. 7;

FIG. 9 is a diagram showing a horse head vitiligo pattern (No. 1).

FIG. 10 is a view showing a vitiligo pattern of a horse head (No. 2).

FIG. 11 is a diagram showing feature names of vitiligo of a head.

FIG. 12 is a flowchart illustrating a process of a curl identification unit 23 in FIG. 1;

FIG. 13 is a diagram showing positions where a spiral pattern is detected.

FIG. 14 is a diagram showing the position and name of the spiral hair of the horse.

FIG. 15 is a flowchart showing a process of a limb vitiligo identification unit 24 in FIG. 1;

FIG. 16 is a diagram showing the extraction region of FIG. 15;

FIG. 17 is a diagram showing feature names of limb vitiligo.

18 is a diagram showing horse names and feature names stored in a database 31 in FIG. 1;

FIG. 19 is a configuration block diagram of an individual identification device showing a second embodiment of the present invention.

[Explanation of symbols]

 DESCRIPTION OF SYMBOLS 10 Image input part 11-13 Camera 20 Image analysis means 30 Horse name output part 31 Database 90 Horse name input part 91 Truth judgment part H Horse

Claims (12)

[Claims] 1. An image input unit for photographing an individual to be identified from different directions and inputting an image of the individual to be identified, and analyzing the image to extract characteristics of the appearance of the individual to be identified. An individual identification device, comprising: an image analysis unit; and an output unit that obtains a name for identifying the individual to be identified based on the characteristic. 2. The image analysis means selects one feature name corresponding to the image for each element from feature names defined and assigned in advance to a variable for each element of the appearance. 2. The individual identification device according to claim 1, wherein the feature is extracted by performing the extraction. 3. A database is provided in which data representing the appearance characteristics of a plurality of individuals using the feature names and the names representing the individuals are stored in association with each other, and the output unit is configured to output the data from the database. 3. The individual according to claim 2, wherein a name of the individual having a characteristic name matching the characteristic name selected by the image analysis means is obtained as a name of the individual to be identified. 4. Identification device. 4. The individual to be identified and the plurality of individuals are animals, and in the image analysis means, the database, and the output unit, the animal's coat color, a pattern of vitiligo of the head, and curly hair are used as the appearance elements. Position or one of the limb vitiligo patterns
4. The individual identification device according to claim 2, wherein at least one is used. 5. The image analysis means extracts at least one of a position of a wound of the animal to be identified or a state of a spot other than a white spot from the image, and the database stores each of the plurality of animals. The name and the position of the wound or the state of the spot of each animal are stored in association with each other, and the output unit searches the database using the position of the wound or the state of the spot extracted from the image, and searches the database. 5. The individual identification device according to claim 4, wherein a name of an animal to be identified is obtained. 6. A name input means for inputting the name of an individual to be identified, an image input unit for photographing the individual to be identified from different directions and inputting an image of the individual to be identified, and analyzing the image. Image analysis means for extracting the appearance characteristics of the individual to be identified, and the names of a plurality of individuals and data representing the appearance characteristics of each individual are stored in association with each other, and are stored via the name input means. A database for searching and outputting the data corresponding to the input name, and comparing the features searched for in the database with the features extracted by the image analysis means to determine the authenticity of the individual to be identified. And a true / false determination unit. 7. The image analysis means selects one feature name corresponding to the image from among feature names predefined and assigned to a variable for each element of the appearance for each element. 7. The individual identification device according to claim 6, wherein the feature is extracted by performing the extraction. 8. The database accumulates data representing the appearance characteristics of the plurality of individuals using the feature names and the names representing the individuals in association with each other. 8. The individual according to claim 7, wherein the feature name retrieved from the database and the feature name selected by the image analysis means are compared to determine the authenticity of the individual to be identified. Identification device. 9. The image analysis unit, the database, and the authenticity determination unit, wherein the individual to be identified and the plurality of individuals are animals, and the animal hair color and head vitiligo pattern are used as the appearance elements. 9. The method according to claim 7, wherein one or more of the position of the trichome and the vitiligo pattern of the limb are used.
An individual identification device as described in the above. 10. The image analysis means according to claim 1, wherein the position of a wound of the animal to be identified or a state of a spot other than a white spot is selected.
One or more extracted from the image, the database accumulates the names of the plurality of animals and the positions of the wounds or the states of the spots of the animals in association with each other, and 10. The individual identification apparatus according to claim 9, wherein the authenticity of the animal to be identified is determined using a position of a wound or a state of a spot extracted from the image. 11. The image input unit according to claim 1, wherein the image input unit is arranged around the individual to be identified, and includes a plurality of cameras that respectively image the individual to be identified.
The individual identification device according to 2, 3, 4, 5, 6, 7, 8, 9, or 10. 12. The apparatus according to claim 1, wherein the image input unit comprises a single camera that moves around the individual to be identified and shoots the individual to be identified.
The individual identification device according to 5, 6, 7, 8, 9 or 10.