Optical nose print registration device and nose print registration method

Abstract

To provide a muzzle pattern registration device and a muzzle pattern registration method which are used for muzzle pattern authentication of an animal, can optically capture a face image and a muzzle pattern image of the animal and process muzzle pattern registration.SOLUTION: A muzzle pattern image acquisition device 110 includes a camera part 111, a muzzle pattern sensor part 113, an operation part 115, a communication module, a processor, and memory. The device performs the acquisition of a muzzle pattern image, a muzzle pattern registration process, and muzzle pattern authentication, displays a face-sight on a screen of an external terminal when capturing a face image, corrects the face-sight so as to match eyes and a nose AN on the face image, displays the nose AN-sight on the screen when capturing a nose AN image, and performs the operation of correcting the nose AN-sight so as to match the philtrum or nostrils on the nose AN image, and the operation of extracting feature point data from the muzzle pattern image.SELECTED DRAWING: Figure 1

Description

[Technical field]

[0001] The present invention relates to a nose print registration device used for animal nose print authentication, which can optically obtain an animal's face image and nose print image and process nose print registration. [Background technology]

[0002] For various purposes, we need to authenticate animals around us to determine whether they are the same animal. For example, if you lose your pet dog, it is very difficult to identify your pet dog among many other dogs of the same species. In addition, it is necessary to verify whether the cows raised by livestock farmers are the same cows whose management numbers have not changed during the breeding and distribution process, but this is difficult to verify simply by their appearance.

[0003] As a way to solve this problem, the method of attaching RFID (Radio Frequency Identification) tags to the ears of cows has been used for a long time, and for pet puppies and cats, it has been common to insert microchips under the skin and record genetic information. In 2013, a companion animal registration system was implemented, and the government will support the registration of microchips and genetic information.

[0004] However, RFID tags are often tampered with without permission and cannot be detected. Microchips can cause side effects and pain to animals, but they have the disadvantage of being easily removed. On the other hand, methods that register genetic information have the disadvantage that they cannot be immediately detected, although they cannot be tampered with.

[0005] Recently, a method has been proposed to authenticate animals using nose prints, i.e. the wrinkles on an animal's nose. Animals have wrinkles on their noses, and these wrinkles are unique to each individual, just like human fingerprints. Nose print authentication is a method of acquiring an image of a nose print and registering the features of the pattern, such as wrinkles, to authenticate the animal.

[0006] The conventional method of acquiring a nose print image for nose print authentication is to directly photograph the nose print using a camera (see Korean Patent Publication No. 10-2019-0066323). However, simply photographing a nose print with a camera is highly affected by not only direct light but also surrounding lighting conditions, making it difficult to acquire a nose print image with a resolution high enough to recognize the wrinkles on the nose print.

[0007] Meanwhile, Korean Patent No. 1706345 presents an optical nose print authentication device using a prism. This method is a direct application of the optical fingerprint acquisition method used in fingerprint recognition. Like fingerprints, nose print patterns can also be divided into ridges and valleys, but optical nose print image acquisition utilizes the fact that the light shining on the nose print performs different optical processing on the ridges and valleys.

[0008] The patent suggests that a prism is used. However, even with the optical nose print image acquisition method proposed in the patent, it is not easy to acquire a nose print image that can recognize wrinkle patterns. This is because an animal's nose is wet, so the valleys of the nose print are filled with water or oil, and optical processing similar to ridges occurs in the valleys of the nose print, making it impossible to distinguish between ridges and valleys, resulting in an overall black or overall light image. [Prior art documents] [Patent documents]

[0009] [Patent Document 1] Korean Patent No. 10-1706345 (Nose print recognition processing device for pet dog management) [Patent Document 2] Korean Patent Publication No. 10-2019-0066323 (Nose Print Photography System) Summary of the Invention [Problem to be solved by the invention]

[0010] SUMMARY OF THE PRESENT EMBODIMENT An object of the present invention is to provide a nose print registration device and method for use in animal nose print authentication, which can optically obtain an animal's face image and nose print image and process nose print registration. [Means for solving the problem]

[0011] In order to achieve the above object, a nose print registration device according to the present invention is connected to an external terminal and can acquire an image of an animal's nose print. The nose print registration device according to the present invention includes a nose print sensor unit, a camera unit, a memory and one or more processors.

[0012] The nose print sensor unit generates a nose print image for a nose print contacting the nose print contact surface, and the camera unit photographs an animal in front to generate at least one selected from a face image and a nose image. Here, the face image is an image of the whole face of the animal, and the nose image is an image of the area around the nose of the animal. A plurality of computer readable commands related to nose print registration are stored in the memory.

[0013] The one or more processors execute commands stored in the memory to perform operations of displaying a face-scale on the screen of the external terminal when the face image is captured, correcting the face-scale to match the eyes and nose on the face image, displaying a nose-scale on the screen when the nose image is captured, correcting the nose-scale to match the philtrum or nostrils on the nose image, and extracting feature point data from the nose pattern image.

[0014] According to another embodiment, the processor may execute the command to generate registration information including the face image, nose image, feature point data, and the identification code, and transmit the registration information to the external terminal to request registration.

[0015] Recognizing and classifying animal patterns According to an embodiment, the memory may store a plurality of face pattern classification information classified according to the geometric arrangement of the eyes and nose of an animal, and a plurality of nose pattern classifications may be stored based on the nose pattern. The processor may further perform an operation of extracting an identification code of a classification to which the corrected face-sight belongs from among the plurality of face pattern classifications and extracting an identification code of a classification to which the corrected nose-sight belongs from among the plurality of nose pattern classifications by executing the command.

[0016] According to another embodiment, the processor may execute the instructions to further perform the operations of recognizing a philtrum in the muzzle image and correcting the muzzle image so that the recognized philtrum is vertically aligned.

[0017] According to still another embodiment, the nose print registration device of the present invention may further include a guide member made of an elastic material, which is disposed higher than the nose print contact surface along the entire or partial area of ​​the upper edge of the nose print contact surface to guide the animal's nose to the nose print contact surface and prevent the animal's nose from being pressed when it comes into contact with the nose print contact surface.

[0018] How to register your nose print The present invention also relates to a nose print registration method of the nose print registration device. The nose print registration method of the present invention includes the steps of: a camera unit takes an image of the entire face of an animal in front to generate a face image, displaying a face-illumination scale on a screen when the face image is taken, and correcting the face-illumination scale so as to match the eyes and nose on the face image; a camera unit takes an image of the area around the nose of the animal to generate a nose image, displaying a nose-illumination scale on a screen when the nose image is taken, and correcting the nose-illumination scale so as to match the philtrum or nostrils on the nose image; and generating a nose print image of the nose print that contacts the nose print sensor unit and extracting feature point data from the nose print image. Effect of the Invention

[0019] The nose print image capturing device of the present invention can capture a nose print image optically by using an optical refractor.

[0020] In addition, the nose print registration device of the present invention is equipped with a camera unit and can generate a face image and a nose image to be registered together with the nose print, and can extract pattern features from the face image or nose image and register them together.The present invention classifies into a plurality of categories based on the geometric pattern features of an animal's face or nose, rather than the nose print features, and by going through a process of primary classification based on the pattern features in the nose print authentication process, it is possible to reduce the time required for nose print authentication and increase authentication efficiency.

[0021] The number of animals, including companion animals, that require nose print recognition can be in the millions or tens of millions, and this solves the inefficiency of simply performing nose print recognition using feature point data extracted from nose print images. For example, when finding an abandoned dog, if a primary classification is performed based on the pattern features and then nose print recognition is performed from the primary classified list, the amount of data that needs to be compared can be significantly reduced.

[0022] In addition, by correcting not only face images and nose images but also nose print images to a certain standard form, image processing errors are reduced when performing nose print recognition or face recognition, and image processing is made easier, thereby increasing the efficiency of nose print recognition.

[0023] The nose print registration device of the present invention uses a guide member to guide the animal's nose to the nose print contact surface during the process of acquiring a nose print image in a contact type, and prevents the animal's nose from being pressed and causing pain. [Brief description of the drawings]

[0024] [Figure 1] 1 is a diagram illustrating a nose print image capturing device according to an embodiment of the present invention. [Diagram 2] FIG. 13 is a diagram showing the configuration of a nose print sensor unit using an optical refractor. [Diagram 3] 1 is a block diagram of a nose print image capturing device according to the present invention; [Figure 4] 2 is a cross-sectional view taken along line xx' in a state where a nose print is in contact with the nose print image capturing device of FIG. 1. FIG. [Diagram 5]2 is a flowchart illustrating a procedure for registering a nose print in the nose print image capturing apparatus according to the present invention. [Figure 6] 13 is a diagram illustrating an example of a face-sight scale displayed on a screen. [Figure 7] 1 is a diagram illustrating an example of a nose-sight scale displayed on a screen. [Figure 8] 1 is a diagram illustrating an example of a nose print image. DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS

[0025] The invention will now be described in detail with reference to the drawings.

[0026] 1 to 3, the nose print image acquiring device 110 of the present invention includes a camera unit 111, a nose print sensor unit 113, an operation unit 115, a communication module 301, a processor 310 and a memory 330, and performs nose print image acquisition, nose print registration and nose print authentication.

[0027] The nose print image capturing device 110 is preferably embodied in a portable type having a portable case 119. The case 119 is preferably a portable type having a handle 119a that allows the user to hold and carry it as shown in FIG. 1, but it may be in the form of a wall-attached type, a desktop type, or a stand type that stands on the bottom. For example, as shown in FIG. 1, the case 119 may be in the shape of a "gun" having a handle 119a on the bottom that is folded in the shape of a "┐". The nose print contact surface 113a of the nose print sensor unit 113 is exposed on one side of the case 119, and a guide member 117 is provided to guide the nose print contact surface 113a to contact the animal's nose print AN.

[0028] The camera unit 111 takes a picture of a subject (animal's face) in front of the camera unit 111 under the control of the processor 310 to generate a digital image. The image generated by the camera unit 111 may include a "face image" in which the entire face of the animal is captured, and / or a "nose image" in which the area around the animal's nose is captured. (b), (c), and (d) of FIG. 6 are examples of face images, and (b), (c), and (d) of FIG. 7 are examples of nose images. As described below, when a user operates and controls the operation unit 115, the processor 310 controls the camera unit 111 to capture an image of a subject in front of the camera unit 111 under that control. The user can capture an image while viewing an image displayed on the screen 131 of the external terminal 130, and check the captured image. The camera unit 111 of the present invention generates a "face image" and a "nose image", but does not generate a nose print image.

[0029] The nose print sensor unit 113 is a contact type nose print image generating device in which the nose print directly contacts the nose print contact surface 113a. The nose print sensor unit 113 may be any device that obtains a nose print image in a contact manner, including an optical method. The nose print sensor unit 113 in Fig. 2 includes an optical refractor 201 with which the nose print directly contacts, an internal light source 203 that emits light for obtaining a nose print image in the optical refractor 201, a lens 205 that focuses the nose print image emitted from the optical refractor 201 on an image sensor 207, and an image sensor 207 that finally generates a digital nose print image, and is controlled by a processor 310.

[0030] In the present invention, the touching of an animal's nose, i.e., a nose print, to the nose print contact surface 113a is referred to as "nose print input". When the animal's nose touches the nose print contact surface 113a, the internal light source 203 of the nose print sensor unit 113 shines light toward the nose print contact surface 113a of the optical refractor 201. The light irradiated by the optical refractor 201 is converted into a nose print image by the nose print through the process of reflection, refraction, and scattering at the nose print contact surface 113a. The nose print image is output through the output surface 201b of the optical refractor 201 and enters the lens 205. In the scattering type nose print sensor unit 113 as shown in FIG. 2, the light output from the internal light source 203 and entering the nose print contact surface 113a at a right angle or an angle smaller than the critical angle for total reflection is converted into a nose print image through other optical processes at the part where the island (AN2) of the nose print touches and the part where the river (AN3) touches. The nose print image is focused on the image sensor 207 by the lens 205. The image sensor 207 converts the nose print image into a digital nose print image and outputs it, thereby obtaining the nose print image of the nose print AN that has come into contact with the nose print contact surface 113a.

[0031] The operation unit 115 is a means for receiving input from a user of a control command to start shooting with the camera unit 111 or the operation of the nose print sensor unit 113, and may be an operation means such as a button or a trigger. The processor 310 executes a preset command in response to the operation of the operation unit 115.

[0032] A guide member 117 is provided on the outer surface of the case 119 at the edge surrounding the nose print contact surface 113a, excluding the lower edge region (upper edge and both side edges). The guide member 117 guides the animal's nose AN so that it can be accurately positioned on the nose print contact surface 113a without deviating from the nose print contact surface 113a. In addition, the guide member 117 is made of an elastic material such as silicone, so that the nose of the animal is not pressed when it touches the nose, preventing the animal from feeling pain.

[0033] 1 and 4, the guide member 117 is formed higher than the muzzle contact surface 113a to prevent the animal's nose AN in contact with the muzzle contact surface 113a from sliding upwards or to either side and leaving the muzzle contact surface 113a. Also, the muzzle image must capture at least the space between the nostrils in the shape of the animal's nose AN, and the upper part of the nose AN must not come off the muzzle contact surface 113a.

[0034] The guide member 117 may be disposed on the upper edge (whole or a part of the area) of the nose print contact surface 113a, and may additionally be provided on the edges (whole or a part) of both sides of the nose print contact surface 113a. For example, Fig. 1 and Fig. 4 show an example in which a hill-shaped guide member 117 with a high center is disposed along the entire upper edge of the nose print contact surface 113a and the upper parts of the edges of both sides. Also, the guide member 117 is formed with an inclined surface 151 toward the nose print contact surface 113a, so that when the animal's nose AN comes into contact with the nose print contact surface 113a, the animal's nose AN can be guided to the nose print contact surface 113a.

[0035] No guide member is provided on the edge of the lower side of the nose print contact surface 113a. In other words, the lower part of the nose print contact surface 113a on the outer surface of the case 119 should be embodied at the same height as the nose print contact surface 113a or at a height slightly lower than the nose print contact surface 113a. As shown in FIG. 4, it is preferable that no guide member is provided on the lower parts of both sides of the nose print contact surface 113a. In other words, it is preferable that the lower parts of both sides of the nose print contact surface 113a on the outer surface of the case 119 should be embodied at the same height as the nose print contact surface 113a or at a height slightly lower than the nose print contact surface 113a. As shown in FIG. 1 and FIG. 4, the lower part and both sides of the nose print contact surface 113a may be embodied at a height slightly higher than the nose print contact surface 113a, but the height of such a structure should not be high enough to prevent the animal's nose AN from slipping or to obstruct its path, like the guide member 117.

[0036] The communication module 301 is a communication means for connecting to the external terminal 130 or a server (not shown) and may be a wired interface such as a Universal Serial Bus (USB) or a wireless interface such as Bluetooth or Wireless LAN. For example, if the communication module 301 is implemented as a USB and the nose print image acquiring device 110 and the external terminal 130 are directly connected through a wired cable, the nose print image acquiring device 110 may receive operating power from the external terminal 130. If the communication module 301 is implemented as a wireless interface, the nose print image acquiring device 110 must have its own operating power source such as a battery.

[0037] The processor 310 executes instructions stored in the memory 330 to control the overall operation of the nose print image acquiring device 110 of the present invention. The processor 310 may be embodied as one or more processors. The processor 310 executes various software programs and / or instruction sets stored in the memory 330 to perform various functions of the nose print image acquiring device 110 and process data.

[0038] The memory 330 stores various information, data, commands, software, etc., including an operating system, required for the operation of the nose print image acquisition device 110. The memory 330 may include at least one of a volatile memory or a non-volatile memory, or a combination thereof, and may include one or more computer-readable storage media. The memory 330 may include at least one of a flash memory type, a hard disk type, a multimedia card micro type, a card-type memory (e.g., SD or XD memory), a random access memory (RAM), a static random access memory (SRAM), a read-only memory (ROM), an electrically erasable programmable read-only memory (EEPROM), a programmable read-only memory (PROM), a magnetic memory, a magnetic disk, and an optical disk. The memory 330 may also correspond to a web storage or a cloud server that performs a storage function on the Internet.

[0039] The memory 330 stores a sensor control module 331, a feature point extraction module 333, a pattern recognition module 335, and a registration processing module 337, which are command sets or software (or applications) for performing nose print registration and / or nose print authentication operations of the present invention. Each of the sensor control module 331, the feature point extraction module 333, the pattern recognition module 335, and the registration processing module 337 operates as one hardware configuration while being executed by the processor 310, and the operation thereof will be described below as if it were a hardware configuration. Meanwhile, these command sets or applications are stored in the memory 330 in advance, or are received from a cloud server and stored.

[0040] The sensor control module 331 is related to the control of the camera unit 111 and the nose print sensor unit 113, and causes the camera unit 111 to generate a face image or a nose image, and causes the nose print sensor unit 113 to generate a nose print image of the nose print AN that comes into contact with the nose print contact surface 113a.

[0041] The feature point extraction module 333 is related to image processing and extracts feature point data from the nose print image generated by the nose print sensor unit 113. The feature points in the nose print image should include not only the morphological characteristics of individual islands (or cells) in an animal's nose print, but also correlation information regarding the morphological arrangement between islands.

[0042] The pattern recognition module 335 is an artificial intelligence algorithm for image processing, and extracts pattern features from a face image or nose image generated by the camera unit 111 using an artificial intelligence algorithm. The pattern features include the geometric arrangement relationship between the two eyes and nose extracted from the face image, the color of the nose extracted from the nose image, the shape of the nose (exterior pattern, pattern and size of the nostrils), and the size of the nose. Therefore, the pattern processed by the pattern recognition module 335 is not a nose print pattern. The nose print pattern is processed by the feature point extraction module 333.

[0043] The geometric arrangement between the two eyes and nose is determined by first checking the positions of the eyes and nose in the face image, then calculating the pixel distance between the eyes and nose, and then calculating the ratio of the distance between the two eyes (d1) to the distance between the two eyes and nose (d2). The external pattern of the nose and nostrils is determined by extracting the outlines of the nose and nostrils from the nose image, and then extracting the pattern features.

[0044] Although animals cannot be authenticated by the features of their patterns, animals can be classified based on the features of their patterns. For example, since the arrangement between the eyes and nose of an animal is slightly different for each animal, animals can be classified according to the shape of the triangle formed by the two eyes and nose of the animal. This classification is useful as the number of registered animals increases since it can reduce the time required for animal authentication, and is useful when other information cannot be input by the user, such as when authenticating a stray dog. The pattern recognition module 335 can assign an identification code classified according to a preset system to the features of the facial pattern and / or nose pattern of the animal.

[0045] According to an embodiment, the pattern recognition module 335 may include a plurality of animal pattern templates or a corresponding classification system to classify the animal to be authenticated. Since the pattern templates are individually assigned identification numbers, when the facial pattern features or nose pattern features of the animal being authenticated match one of the plurality of pattern templates, the animal is mapped and classified by the identification number of the matching template. The pattern templates proposed by the present invention are very simplified animal patterns or nose patterns, and the face-sights G1, G1a and nose-sights G2, G2a described below are also examples of pattern templates.

[0046] The registration processing module 337 performs part of the procedure for registering the target animal to the external terminal 130 or server together with the external terminal 130 by generating registration information required for authenticating the target animal and providing it to the external terminal 130. The registration procedure performed by the registration processing module 337 will be described again below. The registration information provided by the registration processing module 337 to the external terminal 130 includes feature point data generated by the feature point extraction module (333), and may include a nose image and / or a face image generated by the camera unit 111 according to an embodiment. The registration processing module 337 may also include pattern features extracted from the nose image and / or face image by the pattern recognition module 335 (or an identification code classified according to the pattern features) in the registration information.

[0047] How to register an animal's nose print (Figures 5 to 7) In the present invention, the procedure for animal registration is controlled by the registration processing module 337. The processor 310 generates animal registration information by executing each step of Fig. 5 according to the registration processing module 337.

[0048] <Generating face image: S501> When the registration procedure is started, the camera unit 111 captures an image in front of the user. When the camera unit 111 captures an image, the processor 310 provides the video to the external terminal 130 using the communication module 301. At this time, as shown in Fig. 6(a), a face-scope G1 for capturing a face image is displayed on the screen 131 of the external terminal 130.

[0049] Here, the face-scope G1 is a scope that is displayed on the screen 131 when photographing a face image and guides the user to align at least one of the animal's eyes and nose, and is used to ensure basic resolution of the face image. As shown in Fig. 6(b), when photographing an animal's face x1, if the face-scope G1 is displayed on the screen 131 and the animal's eyes and nose are approximately aligned with the face-scope G1, a face image of a certain size and resolution can be ensured.

[0050] Thereafter, if operation unit 115 is operated, camera unit 111 captures an image and generates a facial image.

[0051] <Facial image correction: S503> The pattern recognition module 335 determines whether the eyes and nose of the animal in the face image match the face-scope G1, and if they do not match, it transforms the face-scope G1 to match the eyes and nose of the face image. At this time, the pattern recognition module 335 uses an artificial intelligence algorithm to recognize the eyes and nose of the animal in the face image, and transforms the face-scope G1 to match the recognized eyes and nose.

[0052] FIG. 6 illustrates an example of correcting a face image using a face-scope G1. Even if we try to match the face-scope G1, there are cases where only a partial match is necessary, as in FIG. 6(b). In such a case, the pattern recognition module 335 uses an artificial intelligence algorithm to recognize the eyes and nose of the animal's face x1, and transforms the face-scope G1a to match the recognized eyes and nose, as in FIG. 6(c). If the face-scope G1a that matches the recognized eyes and nose is tilted, as in FIG. 6(c), the pattern recognition module 335 rotates the face image to straighten the tilted face-scope G1a and correct it, as in FIG. 6(d).

[0053] When the tilted face-scope G1a is set up and corrected as shown in FIG. 6(d), the animal-face image has a unified form, facilitating image processing during the animal recognition process.

[0054] <Generating a nose image: S505> When the correction of the face image is completed or there is an additional operation of the operation unit 115, a nose-sight G2 for photographing the nose image is displayed on the screen 131 of the external terminal 130 as shown in FIG. 7(a).

[0055] The nose-sight G2 is a sight that is displayed when photographing a nose image and is guided to match the philtrum and / or nostrils of the animal's nose, and is also used to ensure the basic resolution of the nose image. As shown in Fig. 7(b), when photographing the animal's nose x2 part, the nose-sight G2 is displayed on the screen 131 and the philtrum and nostrils between the animal's nostrils are approximately matched to the nose-sight G2, thereby ensuring a nose image of a certain size and resolution. Another example of the nose-sight may be displayed as two circles corresponding to the two nostrils.

[0056] Thereafter, if operation unit 115 is operated, camera unit 111 generates a nose image.

[0057] <Nose image correction: S507> The pattern recognition module 335 recognizes the animal's nose x2 in the nose image and determines whether the nostrils match the nose-scope G2, and if not, transforms the orientation of the nose-scope G2 to match the philtrum and nostrils.

[0058] FIG. 7 illustrates an example of correcting a nose image using nose-sight G2. When trying to match nose-sight G2 to an animal's nose, it may not be possible to match the direction as shown in FIG. 7(b). In such a case, the pattern recognition module 335 uses an artificial intelligence algorithm to recognize the philtrum and nostrils of the animal's nose x2, and transforms the nose-sight G2a to match the recognized philtrum or nostrils as shown in FIG. 7(c). If the nose-sight G2a matched to the recognized philtrum and nostrils is tilted as shown in FIG. 7(c), the pattern recognition module 335 rotates the nose image to straighten the tilted nose-sight G2a and correct it as shown in FIG. 7(d).

[0059] When the inclined nose-scope G2a is corrected as shown in FIG. 7(d), the animal-nose image has a unified form, which facilitates image processing during the animal recognition process.

[0060] <Extraction of pattern features: S509> Once the nose image has been generated and the nose-sight G2 correction is complete, the pattern recognition module 335 extracts pattern features in the nose image and / or face image.

[0061] For example, the pattern recognition module 335 may have a plurality of face pattern classifications based on the geometric arrangement patterns of the eyes and nose, and may have a plurality of nose pattern classifications based on the nose pattern, each of the plurality of face pattern classifications may use a face-slope representative of the classification, and each of the plurality of nose pattern classifications may use a nose-slope representative of the classification.

[0062] Therefore, when the corrected face-sight G1a is determined in S503, it is possible to determine which of a plurality of preset face pattern classifications the animal's face belongs to. Even if there is no face pattern classification that matches the animal's face, the pattern recognition module 335 will match the most similar classification. When the corrected nose-sight G2a is determined in step S507, it is possible to determine which of a plurality of preset nose pattern classifications the animal's nose pattern belongs to. Even if there is no nose pattern classification that matches the animal's nose pattern, the pattern recognition module 335 will match the most similar classification.

[0063] <Acquisition and correction of nose print image: S511> When the nose image is generated and the correction of the nose-sight G2 is completed or the user operates the operation unit 115, the nose print image acquisition process is started by the sensor control module 331. The sensor control module 331 controls the nose print sensor unit 113 to generate a nose print image of the nose print that contacts the nose print contact surface 113a.

[0064] Once the nose print image is generated, the pattern recognition module 335 recognizes the philtrum and nostrils in the nose print image, and if the philtrum is tilted, corrects the orientation of the nose print image so that the philtrum is vertical. At this time, an artificial intelligence algorithm is used to recognize the philtrum and nostrils in the nose print image.

[0065] 8 illustrates a nose print image, where (a) is the nose print image generated by the camera unit 111, and (b) is the nose print image processed to extract feature point data by the feature point extraction module 333. Correction of the posture of the nose print image may be performed using the nose print image generated by the camera unit 111, or may be performed on the nose print image processed by the feature point extraction module 333 in step S513 described below.

[0066] <Feature data extraction: S513> The feature point extraction module 333 extracts feature point data from the nose print image generated by the nose print sensor unit 113. In order to register a nose print image, it is better to obtain a plurality of nose print images, extract feature point data, and compare them to minimize errors in the feature point data.

[0067] <Create registration information: S515> The registration processing module 337 generates registration information including a nose image and / or face image, a nose print image, feature point data, and pattern classification information (or its identification code) obtained by performing a series of procedures, and provides the registration information to the external terminal 130 to request registration.

[0068] Not only the feature point data but also other registration information can be used in the nose print authentication step. For example, if all pet animals were registered, millions to tens of millions of animals would be registered, and there would be millions of companion dogs alone. If an abandoned dog is found and needs to be authenticated, the feature point data of the nose print extracted from the abandoned dog must be compared with millions of data. At this time, various different registration information is classified in various ways, so classification is performed in the authentication step and the nose print authentication process can be efficiently carried out.

[0069] In the above manner, the registration procedure of the present invention is performed by the nose print image capture device 110. Meanwhile, depending on the embodiment, only one of the face image generation operations in S501 and S503 and the nose image generation operations in S505 and S507 may be performed. In this case, S509 is performed on the generated face image or nose image.

[0070] Although the preferred embodiments of the present invention have been illustrated and described above, it goes without saying that the present invention is not limited to the specific embodiments described above, and various modifications may be made by those having ordinary skill in the art to which the invention pertains without departing from the gist of the present invention as claimed in the claims, and such modifications should not be understood individually from the technical ideas and perspectives of the present invention.

Claims

1. A nose print registration device that is connected to an external terminal and can acquire an image of an animal's nose print, a nose print sensor unit for generating a nose print image for a nose print contacting the nose print contact surface; a camera unit for capturing an image of an animal in front of the camera and generating at least one of a face image and a nose image, the face image being an image of the entire face of the animal, and the nose image being an image of the area around the nose of the animal; a memory storing a plurality of computer readable instructions relating to nose print registration; and one or more processors capable of executing instructions stored in the memory, The memory stores a plurality of face pattern classification information classified according to the geometric arrangement of the eyes and nose of animals, and stores a plurality of nose pattern classifications based on nose patterns, The processor executes the instructions to: displaying a face-perspective on the screen of the external terminal when the face image is photographed, and correcting the face-perspective so as to match the eyes and nose on the face image, and displaying a nose-perspective on the screen when the nose image is photographed, and correcting the nose-perspective so as to match the philtrum or nostrils on the nose image; and performing an operation of extracting feature point data from the nose print image; The processor executes the command to further perform an operation of extracting an identification code of a classification to which the corrected face-sight belongs from among the plurality of face shape classifications, and extracting an identification code of a classification to which the corrected nose-sight belongs from among the plurality of nose shape classifications. Nose print registration device.

2. The processor executes the command to generate registration information including at least one selected from the face image and the nose image, feature point data, and the identification code, and transmits the registration information to the external terminal to request registration.

2. The nose print registration device according to claim 1.

3. The processor executes the instructions to further perform the operations of recognizing a philtrum in the muzzle print image and correcting the muzzle print image so that the recognized philtrum is vertically aligned.

2. The nose print registration device according to claim 1.

4. a guide member made of an elastic material that is disposed higher than the muzzle contact surface along the entire or a part of the upper edge of the muzzle contact surface to guide the animal's nose to the muzzle contact surface and not to press the animal's nose when the animal contacts the muzzle contact surface; 2. The nose print registration device according to claim 1.

5. A method for registering a nose print in a nose print registration device that is connected to an external terminal and has a contact type nose print sensor unit for acquiring an image of an animal's nose print, a step of generating a facial image by capturing an entire face of an animal in front of the camera unit, displaying a face-scope on a screen when capturing the facial image, and correcting the face-scope so that the face-scope matches the eyes and nose on the facial image; a step of displaying a nose-sight on a screen when the camera unit captures an image of the nose of the animal and correcting the nose-sight so that the nose-sight coincides with the philtrum or nostrils on the nose image; generating a nose print image of the nose print contacting the nose print sensor unit and extracting feature point data from the nose print image; storing a plurality of face pattern classification information classified according to the geometric arrangement of the eyes and nose of animals in the memory, and storing a plurality of nose pattern classifications based on the nose pattern; extracting an identification code of a classification to which the corrected face-sight belongs from among the plurality of face pattern classifications, and extracting an identification code of a classification to which the corrected nose-sight belongs from among the plurality of nose pattern classifications, A nose print registration method for a nose print registration device.

6. generating registration information including at least one selected from the face image and the nose image, feature point data, and the identification code, and transmitting the registration information to the external terminal through a communication module to request registration; 6. A method for registering a nose print in the nose print registration device according to claim 5.

7. recognizing a philtrum in the muzzle image and correcting the muzzle image so that the recognized philtrum is vertically aligned.

6. A method for registering a nose print in the nose print registration device according to claim 5.

Images