Image tagging-based pet photo classification method and system
The image tagging-based system addresses pet identification and visit management in facilities by using machine learning algorithms to classify pet photos and share them with owners, improving facility visit management and owner communication.
Patent Information
- Application Number
- JP2025546738
- Authority / Receiving Office
- JP · JP
- Patent Type
- Applications
- Current Assignee / Owner
- Priority Date
- 2023-03-23
- Filing Date
- 2024-03-08
- Publication Date
- 2026-02-19
AI Technical Summary
There is a challenge in distinguishing which pet belongs to whom in pet care facilities and a lack of efficient information management and sharing of pet care facility visits for pet owners.
An image tagging-based method and system that uses a few-shot learning algorithm and Siamese neural networks to classify pet photos, determine pet visits, and provide images to owners, utilizing a communication unit, processor, and storage unit to manage pet and owner information and facilitate reservations.
Enables accurate classification and sharing of pet photos within pet care facilities, ensuring efficient management and communication of pet visits to owners, enhancing user convenience.
Smart Images
Figure 2026505899000001_ABST
Abstract
Description
[Technical Field]
[0001] The present invention relates to a method and system for classifying photos, and more particularly to a method and system for classifying pet photos based on image tagging. [Background technology]
[0002] As industry develops, the structure of society changes, and with the aging population, the trend toward nuclear families, and the rapid increase in people living alone, the number of households keeping pets is increasing.
[0003] As more and more households treat their pets like members of the family, the number of people using pet care facilities (such as pet-only kindergartens, pet hotels, and pet salons) that provide care for their pets has also increased sharply in recent years.
[0004] In such pet care facilities, multiple pets spend time together, and in the case of pets of the same species, it can be difficult to distinguish which pet belongs to whom.
[0005] Furthermore, there is no platform that provides users (pet owners) with efficient information management and sharing of pets using pet care facilities, so solutions to this issue need to be explored. Summary of the Invention [Problem to be solved by the invention]
[0006] Therefore, the present invention has been made in consideration of the above problems, and an object of the present invention is to provide a method and system for classifying pet photos that determines whether or not a pet is visiting (attending) a pet care facility based on on-site photos of the pets visiting the facility, classifies each pet that appears in a group photo taken within the pet care facility, and provides a service of forwarding the photos to each pet's owner. [Means for solving the problem]
[0007] To achieve the above-mentioned objective, one embodiment of the present invention provides an image tagging-based pet photo classification method, which includes the steps of: a system registering pet (Pet) information and pet owner information received from a user terminal; a system cooperating with the user terminal to make a reservation for use of a pet care facility based on the pet information and pet owner information; and a system determining whether the pet will visit the pet care facility based on a photo of the pet taken at the pet care facility and the reservation information registered during the reservation process.
[0008] The step of determining whether or not a pet care facility has been visited may determine, when a photograph of the pet taken at the pet care facility is uploaded, whether or not the pet has visited the pet care facility based on the reservation information registered during the reservation process and the uploaded photograph of the pet.
[0009] The step of making a reservation can only be performed if the pet owner logged in to a specific user terminal has registered (purchased) a voucher for the pet care facility, and the step of determining whether or not to visit the pet care facility can be performed by automatically deducting the number of times the voucher registered (purchased) by the pet owner on the user terminal can be used if the visit to the pet care facility is approved based on a photograph of the pet taken at the pet care facility.
[0010] The system may then store, via a photograph of the pet taken on-site at the pet care facility, a tagged photo of the pet's name, selected from a visit list or extracted from reservation information, when the pet is approved to visit the pet care facility.
[0011] The system may utilize a few-shot learning algorithm and Siamese neural networks to perform object detection of a specific pet from a photograph or captured image containing multiple pets and extract an image of the specific pet.
[0012] When the system utilizes a Few Shot Learning algorithm and Siamese Neural Networks, it may apply photos of the pet included in the information of the pet to be registered or photos of the pet taken on-site at the pet care facility as a support set, and apply photos or captured images of multiple pets as a query set.
[0013] In addition, the image tagging-based pet photo classification method according to one embodiment of the present invention may further include a step in which, when photos or captured images of multiple pets are obtained from inside a pet care facility, the system individually extracts the image of each pet and maps it to the most similar group within the support set, and classifies each pet image by pet according to the mapping result.
[0014] The images of the classified pets may then be filtered and stored based on the pet care facility usage time information for each pet.
[0015] In addition, the image tagging-based pet photo classification method according to one embodiment of the present invention may further include a step in which, when an owner of a particular pet requests images of the pet captured from inside the pet care facility via a user terminal, the system transfers the filtered and saved images of the pet to the user terminal.
[0016] Meanwhile, an image tagging-based pet photo classification system according to another embodiment of the present invention includes a processor that registers pet information and pet owner information received from a user terminal, cooperates with the user terminal, and makes a reservation for use of a pet care facility based on the pet information and pet owner information, and determines whether the pet will visit the pet care facility based on photos of the pet taken at the pet care facility and the reservation information registered during the reservation process, and a storage unit that stores the registered pet information and pet owner information, and the reservation information registered during the reservation process. [Effects of the Invention]
[0017] As described above, according to the present invention, it is possible to provide a service that determines whether a pet has visited (attended) a pet care facility based on on-site photographs of the pets that have visited the facility, classifies each pet included in a group photograph taken within the pet care facility, and forwards the photographs to each pet's owner, thereby providing convenience to pet owners. [Brief explanation of the drawings]
[0018] [Figure 1] 1 is a diagram illustrating the configuration of a pet photo classification system according to an embodiment of the present invention; [Figure 2] 1 is a flowchart illustrating a method for classifying pet photos according to an embodiment of the present invention. [Figure 3] 10 is a flowchart illustrating in more detail a process of determining whether a pet has visited a pet care facility according to an embodiment of the present invention. [Figure 4] 10A to 10C are diagrams illustrating a process of extracting a pet from a pet photo or image according to an embodiment of the present invention. [Figure 5] 10A to 10C are diagrams illustrating a process of extracting a pet from a pet photo or image according to an embodiment of the present invention. [Figure 6] 10A and 10B are diagrams illustrating a process of transmitting tagged images to a pet owner according to an embodiment of the present invention; DETAILED DESCRIPTION OF THE INVENTION
[0019] The invention will now be explained in more detail with reference to the drawings. FIG. 1 is a diagram illustrating the configuration of a pet photo classification system (hereinafter referred to as "system") according to an embodiment of the present invention.
[0020] The system of this embodiment can determine whether or not a pet (Pet) has visited (attended) a pet care facility based on on-site photographs of the pet, classify each pet appearing in a group photograph taken within the pet care facility, and provide a service of forwarding the photograph to each pet's owner.
[0021] To that end, the system may include a communication unit 110, a processor 120, and a storage unit 130.
[0022] The communication unit 110 may allow the system to cooperate with a user terminal (for example, a smartphone, a tablet PC, a laptop computer, etc.) used by the pet owner, and may collect and provide information necessary for the processor 120 to operate.
[0023] The storage unit 130 is a storage medium that stores data and programs required for the processor 120 to operate.
[0024] A processor 120 is provided to handle system matters.
[0025] For example, the processor 120 may determine whether a pet has visited (attended) the pet care facility based on on-site photographs of the pet that have visited the pet care facility, and automatically deduct the number of times a pet care facility voucher has been used.
[0026] Specifically, the processor 120 may register information about the pet and the pet owner provided from the user terminal and store the information in the storage unit 130 .
[0027] The processor 120 may then make a reservation for use of the pet care facility based on the pet information and the pet owner information, and may determine whether or not the pet will visit the pet care facility based on a photograph of the pet taken at the pet care facility and the reservation information registered during the reservation process.
[0028] In this case, when a pet is permitted to visit the pet care facility through a photograph of the pet taken at the pet care facility, the processor 120 may store the photograph in the storage unit 130 with the name of the pet selected from the visit list or extracted from the reservation information tagged on the photograph.
[0029] The processor 120 may classify each pet appearing in a group photo taken within the pet care facility and transfer the classification to a user terminal used by each pet's owner.
[0030] Specifically, when a photograph or captured image showing multiple pets inside a pet care facility is obtained, the processor 120 may extract the image of each pet individually, map it to the most similar group within the support set, and, according to the mapping result, classify and store the image of each pet by pet.
[0031] The processor 120 may then transfer the stored images of a particular pet to a user terminal when the owner of that pet requests images of that pet captured from within the pet care facility via the user terminal.
[0032] FIG. 2 is a flowchart illustrating a method for classifying pet photos according to an embodiment of the present invention.
[0033] The pet photo classification method according to this embodiment may be performed by the system described above with reference to FIG.
[0034] Referring to FIG. 2, the method for classifying pet photos may register information about pets and pet owners provided from a user terminal via the system (S210).
[0035] The method of classifying pet photos may involve making a reservation for use of a pet care facility based on information about the pet and the pet's owner (S220), and determining whether the pet has visited the pet care facility based on photos of the pet taken at the pet care facility and reservation information registered during the reservation process (S230).
[0036] Here, the system may store a photograph of a pet taken on-site at the pet care facility, tagging the pet's name, selected from a visit list or extracted from reservation information, with the pet's name when the pet is approved to visit the pet care facility.
[0037] On the other hand, in the pet photo classification method, when photos or captured images of multiple pets are obtained from inside the pet care facility (S240-YES), the images of each pet are individually extracted and mapped to the most similar group within the support set (S250), and depending on the mapping results, the images of each pet may be classified by pet and stored (S260).
[0038] The pet photo classification method may include, when an owner of a particular pet requests an image of the pet acquired from inside the pet care facility via a user terminal, transferring the stored image of the pet to the user terminal (S270).
[0039] FIG. 3 is a flowchart illustrating in more detail the process of determining whether a pet has visited a pet care facility according to one embodiment of the present invention.
[0040] Each pet owner must acquire (purchase) a voucher (e.g., a mobile ticket) for use at a pet care facility before connecting to the system to make a reservation for use at the pet care facility. That is, a pet owner logged into a particular user terminal may make a reservation only if the pet care facility has registered (purchased) the voucher.
[0041] To this end, the system can sell vouchers (e.g., mobile tickets) for use at pet care facilities to each pet owner, who may connect to the server via a user terminal and purchase vouchers for a particular pet care facility.
[0042] An owner of a particular pet may use a user terminal to purchase a voucher for use at a particular pet care facility, then connect to the system to make a reservation for use at the particular pet care facility (S310), and then visit the pet care facility on the date and time of the reservation and register a visit record (attendance record) (S320).
[0043] Here, the visit record book can be registered at the reserved pet care facility using a user terminal, by launching an application provided by the system, and taking a photo of the pet at the pet care facility (S330), uploading it, and then carrying out the pet care facility visit confirmation (authentication) procedure.
[0044] When a photograph of a pet taken on-site at a pet care facility is uploaded, the system may determine whether or not the pet will visit the pet care facility based on the reservation information registered during the reservation process and the uploaded photograph of the pet, and if the visit to the pet care facility is approved based on the photograph of the pet taken on-site at the pet care facility, the system may automatically deduct the number of uses of the voucher registered (purchased) by the pet owner on the user terminal (S340).
[0045] The system may then tag and store the photograph of the pet taken at the pet care facility with the name of the pet selected from the visit list or extracted from the reservation information when the pet is approved to visit the pet care facility via a photograph of the pet taken at the pet care facility (S350).
[0046] FIG. 4A is a diagram illustrating how each pet is object detected from a photo containing multiple pets, FIG. 4B is a diagram illustrating photos of each pet extracted through object detection, and FIG. 5 is a diagram illustrating the process of extracting features from photos of pets included in information on previously registered pets or photos of pets taken on-site at a pet care facility, which are applied as a support set, comparing them with the features of photos of each pet, which are applied as a query set, and tagging the most similar pet.
[0047] When photographs or captured images of multiple pets are obtained from inside the pet care facility, the system may extract each pet image individually, apply them as a query set, map them to the most similar group in the support set, and classify each pet image by pet according to the mapping results.
[0048] In this case, the system may utilize a few-shot learning algorithm and Siamese neural networks to perform object detection of a specific pet from a photograph or captured image containing multiple pets and extract an image of the specific pet, as shown in FIG. 4A.
[0049] When utilizing a human-shot learning algorithm and a Siamese neural network, the system may apply photos of the pet included in the information of the pet being registered or photos of the pet taken on-site at the pet care facility as a support set, and apply photos or captured images of multiple pets as a query set.
[0050] That is, the system may apply photos of pets enrolled in a human-shot learning algorithm and a Siamese neural network as a support set, and extract the enrolled pets from photos or captured images of multiple pets that fit into the query set.
[0051] Here, the classified pet images may be filtered and stored based on the pet care facility usage time information for each pet.
[0052] That is, according to the mapping results, the system may filter the results of categorizing each pet's image based on the time each pet used the pet care facility, and if there is a result that categorizes pet A's image during a time when pet A did not use the pet care facility, it may filter (remove) it.
[0053] FIG. 6 is a diagram illustrating a process of transmitting tagged images to a pet owner according to an embodiment of the present invention.
[0054] The system may acquire photographs or capture images of a particular pet interacting with other pets while staying at the pet care facility. The system may set various conditions for acquiring photographs or capture images depending on the installation environment of security cameras and other cameras installed at the pet care facility, and may monitor the pet's activities and acquire photos of the pet.
[0055] Then, as described above with reference to Figures 4 and 5, when photos or captured images of pets are obtained, the system may extract each pet image individually and map it to the most similar group within the support set, and according to the mapping results, each pet image may be classified by pet.
[0056] Thereafter, when an owner of a particular pet requests, via a user terminal, an image of that pet captured from inside the pet care facility, the system may transfer the filtered and saved image of the pet to the user terminal.
[0057] For example, as shown in Figure 6, when a pet owner composes a message via a user terminal, by clicking on a photo icon displayed on the screen, the system may automatically load (transfer) only the owner's pet photo from among the pet photos acquired from the pet care facility and display it on the screen.
[0058] Meanwhile, the technical concept of the present invention may also be applied to a computer-readable recording medium incorporating a computer program that causes the functions of the apparatus and method according to the present embodiments to be performed. The technical concept of various embodiments of the present invention may be realized in the form of computer-readable code recorded on a computer-readable recording medium. The computer-readable recording medium may be any data storage device that can be read by a computer and that can store data. For example, the computer-readable recording medium may be a ROM, RAM, CD-ROM, magnetic tape, floppy disk, optical disk, hard disk drive, etc. The computer-readable code or program stored on the computer-readable recording medium may be transmitted via a network connected between computers.
[0059] Although the preferred embodiments of the present invention have been described in detail above with reference to the accompanying drawings, the present invention is not limited to the above embodiments. It is clear that a person skilled in the art to which the present invention pertains can conceive of various modifications or alterations within the scope of the technical spirit described in the claims, and it is understood that these also naturally fall within the technical scope of the present invention.
Claims
1. A step in which the system registers information about the pet (Pet) and information about the pet's owner received from the user terminal; The system cooperates with the user terminal to make a reservation for use of the pet care facility based on information about the pet and the pet owner; the system determines whether the pet has visited the pet care facility based on a photograph of the pet taken at the pet care facility and reservation information registered during the reservation process; An image tagging-based pet photo classification method including.
2. The steps to determine whether or not to visit a pet care facility are: The image tagging-based pet photo classification method of claim 1, characterized in that when a photo of a pet taken at a pet care facility is uploaded, it is determined whether the pet has visited the pet care facility based on the reservation information registered during the reservation process and the uploaded photo of the pet.
3. The steps to make a reservation are: Reservations can only be made when a pet owner who logs in to a specific user terminal registers (purchases) a ticket for a pet care facility. The steps to determine whether or not to visit a pet care facility are: The image tagging-based pet photo classification method of claim 2, characterized in that when a visit to a pet care facility is approved based on a photo of the pet taken at the pet care facility, the number of uses of a voucher registered (purchased) by the pet owner on a user terminal is automatically deducted.
4. The system is 4. The method of claim 3, wherein when a pet is admitted to a pet care facility through a photograph of the pet taken on-site, the photograph is tagged with the name of the pet selected from a visit list or extracted from reservation information and saved.
5. The system is The image tagging-based pet photo classification method according to claim 2, characterized in that it utilizes a few-shot learning algorithm and Siamese neural networks to perform object detection of a specific pet from photos or captured images containing multiple pets and extract images of the specific pet.
6. The system is When using a few-shot learning algorithm and Siamese neural networks, photos of the pet included in the registered pet information or photos of the pet taken at the pet care facility are used as a support set. The image tagging-based pet photo classification method according to claim 5, wherein photos or captured images of a plurality of pets are applied as a query set.
7. 7. The image tagging-based pet photo classification method of claim 6, further comprising the steps of: when photographs or captured images of multiple pets are obtained from inside a pet care facility, the system extracts each pet image individually, maps it to its most similar group within the support set, and classifies each pet image by pet according to the mapping result.
8. Categorised pet images are The method for categorizing pet photos based on image tagging according to claim 7, wherein the photos are filtered and stored based on the usage time information of the pet care facility for each pet.
9. 10. The image tagging-based pet photo classification method of claim 8, further comprising the step of the system transferring the filtered and saved pet images to the user terminal when an owner of a particular pet requests images of that pet captured from inside the pet care facility via the user terminal.
10. a processor that registers information about the pet (Pet) and the pet's owner received from the user terminal, cooperates with the user terminal, and makes a reservation for use of a pet care facility based on the information about the pet and the pet's owner, and determines whether or not the pet will visit the pet care facility based on a photo of the pet taken at the pet care facility and the reservation information registered during the reservation process; a storage unit for storing information on the registered pet and pet owner, and reservation information registered during the reservation process; An image tagging-based pet photo classification system.