Ai-based companion animal missing response system

Abstract

A companion animal missing response system according to an embodiment may comprise: a companion animal wearable device worn on a companion animal and including a GPS module for generating GPS data of the companion animal; a first user terminal, which is a terminal of a guardian of the companion animal, for receiving the GPS data from the companion animal wearable device; an electronic device for generating a digital flyer upon the companion animal going missing; and a second user terminal for receiving, from the electronic device, a notification including the digital flyer. The electronic device may receive, through the first user terminal, GPS data at a missing time point of the companion animal and image data obtained by photographing the companion animal before the missing time point, generate the digital flyer by using a first artificial intelligence model on the basis of the GPS data at the missing time point and the image data, and determine, using a second artificial intelligence model, the second user terminal to which a notification including the digital flyer is to be transmitted.

Description

AI-based pet loss response system

[0001] The disclosure below relates to a pet loss response system, and more specifically, to a pet loss response system based on AI technology.

[0002] Lost pets are a common problem worldwide, and a swift response is essential, especially when a pet becomes displaced due to an unexpected situation outdoors. Generally, responding within three hours of a lost pet is crucial, known as the "golden hour." Failure to take appropriate action within this timeframe increases the likelihood that the pet will be lost far away or end up in a shelter. In the worst-case scenario, it could lead to a traffic accident or exposure to a dangerous environment. Therefore, technologies that can quickly and efficiently track pets within this golden hour and encourage cooperation from pet owners and the local community are in demand.

[0003] To this end, community-based network technologies are being utilized to respond to lost pets. Crowdsourcing-based missing pet reporting systems send real-time notifications to users in the area where pets have been found, encouraging them to report sightings. Recently, the application of wide-area location tracking technology utilizing the LoRa (Long Range) network has enabled more reliable location tracking not only in cities but also in suburban and rural areas. This combination of location tracking technology, network technology, and community-based response systems is increasingly advancing technological solutions to address lost pets.

[0004] <Previous literature>

[0005] 1. Korean Patent No. 10-2724313, Patent No. 10-2695386 (August 9, 2024)

[0006] 2. Korean Patent No. 10-2693018 (August 2, 2024)

[0007] Currently, the leading technologies for pet loss prevention are GPS and Bluetooth-based location tracking devices. GPS triggers are widely used because they can track a pet's real-time location via satellite signals. However, they suffer from high battery consumption and difficulty receiving signals indoors. Bluetooth-based location tracking technology is advantageous for short-distance tracking, but it can be difficult to immediately confirm a pet's location once it leaves the signal range.

[0008] Most photos provided by pet owners when their pets are lost are taken in the past, and are likely to differ from their actual appearance at the time of loss. In particular, external features can change due to changes in fur growth, grooming, seasonal changes, clothing, weight gain, or loss. These changes make it difficult for witnesses to accurately identify their pets based solely on photos in digital flyers, necessitating a more effective identification method.

[0009] Embodiments of the present disclosure are intended to address these needs and provide a system for responding to lost pets using AI technology.

[0010] However, these tasks are exemplary and do not limit the scope of the present disclosure.

[0011] A system for responding to lost pets according to one embodiment may include: a pet wearable device that is worn on a pet and includes a GPS module that generates GPS data of the pet; a first user terminal that serves as a terminal of a guardian of the pet and receives the GPS data from the pet wearable device; an electronic device that generates a digital leaflet when the pet is lost; and a second user terminal that receives a notification including the digital leaflet from the electronic device. The electronic device may receive, through the first user terminal, GPS data at the time of loss of the pet and image data taken of the pet before the time of loss, and may generate the digital leaflet based on the GPS data at the time of loss and the image data using a first artificial intelligence model, and may determine the second user terminal to which a notification including the digital leaflet will be transmitted using a second artificial intelligence model.

[0012] According to one embodiment, the first artificial intelligence model may compare a timestamp included in GPS data at the time of loss with a timestamp included in the image data to generate real-time image data of the pet based on the image data, and the second artificial intelligence model may infer real-time location data of the pet and adaptively determine the second user terminal to which the notification will be sent by considering the real-time location data and the elapsed time since the pet was lost.

[0013] According to one embodiment, the first artificial intelligence model may include an image encoder that outputs a feature map for the external features of the companion animal based on the image data; a time embedding module that embeds a time difference between GPS data at the time of loss and the image data based on the two data; a fusion module that combines the feature map and the time difference embedding to output a fused feature map; and an image decoder that outputs real-time image data of the companion animal based on the fused feature map. The electronic device may generate the digital flyer including the real-time image data of the companion animal.

[0014] According to one embodiment, the fusion module may be characterized by generating γ parameters and β parameters of the same dimension as the feature map based on the time difference embedding, and applying scale adjustment or shift adjustment for each channel of the feature map based on the γ parameters and the β parameters.

[0015] According to one embodiment, the pet wearable device may be characterized by including only a short-range communication module instead of a mobile communication module.

[0016] According to one embodiment, the second artificial intelligence model may include a preprocessing module that outputs initial location data, the elapsed time since the loss, and a condition vector based on GPS data at the time of loss, behavioral pattern information of the pet, and surrounding environment information; a simulation-based location inference module that outputs real-time location data of the pet based on the initial location data, the elapsed time since the loss, and the condition vector; and a notification area determination module that outputs a notification area and a notification transmission target based on the real-time location data and the elapsed time since the loss.

[0017] In one embodiment, the real-time location data of the companion animal may be a weighted sum of the companion animal's predicted location derived from a dynamic prediction network and a correction value generated from a rule-based simulation. The rule-based simulation may include a random walk reflecting a certain level of randomness, a destination directionality reflecting the companion animal's likelihood of moving to a specific destination, and obstacle avoidance rules that avoid physical constraints in the surrounding environment.

[0018] According to one embodiment, the notification area determination module determines a first area included within a first radius based on real-time location data of the pet as a notification area when the time elapsed since the loss is less than a first threshold, and determines a first group of recipients located within the first area as a notification transmission target, and determines a second area included within a second radius based on real-time location data of the pet as a notification area when the time elapsed since the loss is less than a second threshold, and determines a second group of recipients located within the second area as a notification transmission target, wherein the first threshold is less than the second threshold, the first radius is less than the second radius, and the notification transmission target included in the first group of recipients may be less than or equal to the notification transmission target included in the second group of recipients.

[0019] According to one embodiment, the electronic device may be characterized in that it applies a filter suitable for the surrounding environment to the real-time image data based on the real-time location data of the companion animal.

[0020] According to one embodiment, a method of operating an electronic device may include the steps of: receiving, through a first user terminal, which is a terminal of a guardian of a companion animal, GPS data at the time of loss of the companion animal and image data taken of the companion animal before the time of loss; generating the digital leaflet based on the GPS data at the time of loss and the image data using a first artificial intelligence model; and determining, using a second artificial intelligence model, a second user terminal to which a notification including the digital leaflet will be transmitted. The GPS data at the time of loss of the companion animal may be generated from a companion animal wearable device that is worn on the companion animal and includes only a short-range communication module instead of a mobile communication module.

[0021] The present invention enables more accurate and rapid location tracking in the event of a lost pet, significantly improving rescue rates within the golden hour. By addressing the shortcomings of GPS and Bluetooth-based location tracking technologies, the present invention broadens the scope of real-time pet location tracking and provides notifications to nearby users based on the elapsed time since the pet was lost, enabling pet owners to quickly locate and rescue their pets.

[0022] The present invention improves the image of a pet in a digital flyer, allowing witnesses to more easily identify a lost animal. To compensate for the discrepancy between the captured image and the actual appearance at the time of loss, dynamic image generation technology and a filtering system that reflect factors such as grooming, seasonal changes, and body shape can be utilized to provide more accurate appearance information. Utilizing these precisely generated pet images not only increases the efficiency of the owner's search, but also improves the accuracy of witness reports, maximizing the likelihood of a pet's rescue.

[0023] FIG. 1 is a drawing for explaining a pet loss response system according to one embodiment.

[0024] FIG. 2 is a diagram illustrating AI used in a pet loss response system according to one embodiment.

[0025] FIG. 3 is a flowchart of an operation method of a pet loss response system according to one embodiment.

[0026] FIG. 4 is a diagram for explaining a first artificial intelligence model according to one embodiment.

[0027] FIG. 5 is an example of a digital flyer generated using a first artificial intelligence model according to one embodiment.

[0028] FIG. 6 is a diagram for explaining a second artificial intelligence model according to one embodiment.

[0029] FIG. 7 is a diagram illustrating a notification area and a notification transmission target in which a notification including a digital flyer is provided according to one embodiment.

[0030] Specific structural or functional descriptions of the embodiments are disclosed for illustrative purposes only and may be modified and implemented in various forms. Therefore, the actual implementation is not limited to the specific embodiments disclosed, and the scope of this specification includes modifications, equivalents, or alternatives within the technical concepts described in the embodiments.

[0031] Although terms such as "first" or "second" may be used to describe various components, these terms should be interpreted solely to distinguish one component from another. For example, a first component may be referred to as a second component, and similarly, a second component may also be referred to as a first component.

[0032] When it is said that a component is "connected" to another component, it should be understood that it may be directly connected or connected to that other component, but there may also be other components in between.

[0033] Singular expressions include plural expressions unless the context clearly dictates otherwise. In this document, phrases such as "A or B", "at least one of A and B", "at least one of A or B", "A, B, or C", "at least one of A, B, and C", and "at least one of A, B, or C" can each include any one of the items listed together in that phrase, or all possible combinations thereof. In this specification, it should be understood that the terms "comprises" or "has" and the like are intended to specify the presence of a described feature, number, step, operation, component, part, or combination thereof, but do not exclude the presence or addition of one or more other features, numbers, steps, operations, components, parts, or combinations thereof.

[0034] Unless otherwise defined, all terms used herein, including technical or scientific terms, have the same meaning as commonly understood by a person of ordinary skill in the art. Terms defined in commonly used dictionaries should be interpreted to have a meaning consistent with their meaning in the context of the relevant technology, and will not be interpreted in an idealized or overly formal sense unless explicitly defined herein.

[0035] The term "module" as used herein may include a unit implemented in hardware, software, or firmware, and may be used interchangeably with terms such as logic, logic block, component, or circuit. A module may be an integral component, or a minimum unit or portion of such a component that performs one or more functions. For example, according to one embodiment, a module may be implemented in the form of an application-specific integrated circuit (ASIC).

[0036] The term "~part" as used in this document refers to a software or hardware component such as an FPGA or ASIC, and the "~part" performs certain roles. However, the "~part" is not limited to software or hardware. The "~part" may be configured to reside on an addressable storage medium and may be configured to execute one or more processors. For example, the "~part" may include components such as software components, object-oriented software components, class components, and task components, as well as processes, functions, attributes, procedures, subroutines, segments of program code, drivers, firmware, microcode, circuits, data, databases, data structures, tables, arrays, and variables. The functionality provided within the components and "~parts" may be combined into a smaller number of components and "~parts" or further separated into additional components and "~parts." Furthermore, the components and "~parts" may be implemented to execute one or more CPUs within a device or a secure multimedia card. Additionally, '~bu' may include one or more processors.

[0037] Hereinafter, embodiments will be described in detail with reference to the attached drawings. In the description with reference to the attached drawings, identical components are assigned the same reference numerals regardless of the drawing numbers, and redundant descriptions thereof will be omitted.

[0038]

[0039] FIG. 1 is a drawing for explaining a pet loss response system according to one embodiment.

[0040] Referring to Figure 1, a schematic diagram of a pet loss response system (10) can be seen.

[0041] The pet loss response system (hereinafter referred to as system (10)) may be a system built to rescue a lost pet within the golden time.

[0042] The system (10) may utilize an AI model (e.g., 230) to respond to lost pets. The system (10) may utilize the AI ​​model (230) to generate real-time image data of the pet based on existing images of the pet, generate a digital flyer (e.g., 501 of FIG. 5) reflecting the real-time image data, and adaptively determine the target to which the digital flyer should be sent by considering the elapsed time since the pet was lost (e.g., see FIG. 7).

[0043] The system (10) may include a first user terminal (100), a companion animal wearable device (101), an electronic device (200), and a second user terminal (300). The first user terminal (100), the electronic device (200), and the second user terminal (300) may be connected via a network (11) (e.g., a mobile radio communication network, a local area network (LAN), a wide area network (WAN), a value added network (VAN), a satellite communication network, or a combination thereof). The first user terminal (100), the electronic device (200), and the second user terminal (300) can communicate with each other using a wired communication method or a wireless communication method (e.g., wireless LAN (WiFi), Bluetooth, Bluetooth low energy, ZigBee, WFD (WiFi direct), UWB (ultra wide band), infrared communication (IrDA, infrared data association), NFC (near field communication)).

[0044] The first user terminal (100) may be a terminal of a pet owner. The first user terminal (100) may include a memory (110) and a processor (120). The memory (110) may store various data used (or collected) by at least one component (e.g., processor (120)) of the first user terminal (100). The processor (120) (e.g., application processor) may access the memory (110) and execute one or more instructions.

[0045] A companion animal wearable device (101) may be a device worn by a companion animal. The companion animal wearable device (101) may be implemented as a collar, harness, tag, or band, but is not limited thereto. The companion animal wearable device (101) may include a GPS module that generates companion animal GPS data. The companion animal wearable device (101) may include only a short-range communication module instead of a mobile communication module. The companion animal GPS data generated by the GPS module may be transmitted to a first user terminal (100) via the short-range communication module.

[0046] An electronic device (200) (e.g., a server) can generate a digital flyer for a companion animal and determine a target (e.g., a second user terminal (300)) to which a notification including the digital flyer will be sent. The electronic device (200) (e.g., a server) can include a memory (210), a processor (220), and an artificial intelligence model (230). The memory (210) can store various data used (or collected) by at least one component (e.g., the processor (220)) of the electronic device (200). The processor (220) (e.g., an application processor) can access the memory (210) and execute one or more instructions. The artificial intelligence model (230) can be stored in the electronic device (200) after learning is completed. The artificial intelligence model (230) will be described in detail with reference to FIG. 2.

[0047] The second user terminal (300) may be a target that receives a notification including a digital flyer from the electronic device (200). The second user terminal (300) may be a user terminal having an application of a business operator operating a pet loss response system (10) installed, a terminal of an animal hospital affiliated with the business operator operating the system (10), or a terminal of an animal protection organization affiliated with the business operator operating the system (10). The second user terminal (300) may include a memory (310) and a processor (320). The memory (310) may store various data used (or collected) by at least one component (e.g., the processor (320)) of the second user terminal (300). The processor (320) (e.g., the application processor) may access the memory (310) and execute one or more instructions.

[0048] Below, the AI ​​model (230) will be described in detail. The AI ​​model (230) can generate a real-time image of a pet to be used in a digital flyer, and derive a notification area and notification transmission target where the digital flyer will be sent.

[0049] Figure 2 is a diagram for explaining AI used in a pet loss response system according to one embodiment.

[0050] The electronic device (200) can utilize multiple artificial intelligence models (230). The artificial intelligence models (230) may include a first artificial intelligence model (231) and a second artificial intelligence model (232). The first artificial intelligence model (231) and the second artificial intelligence model (232) may be trained based on different learning data.

[0051] The first artificial intelligence model (231) can utilize GPS data at the time of the pet's loss and image data taken of the pet before the loss. The first artificial intelligence model (231) can generate real-time image data of the pet by comparing the timestamp included in the GPS data at the time of the loss with the timestamp included in the image data before the image was lost. The first artificial intelligence model (231) will be described in detail with reference to FIG. 4.

[0052] The second artificial intelligence model (232) may utilize GPS data at the time of the pet's loss, information on the pet's behavioral patterns, and information on the surrounding environment. The second artificial intelligence model (232) may utilize the above data or information to determine the second user terminal (300) to which a notification including a digital flyer will be sent. The second artificial intelligence model (232) may first infer the pet's real-time location data, and then adaptively determine the second user terminal to which the notification will be sent based on the real-time location data and the elapsed time since the pet was lost. The second artificial intelligence model (232) will be described in detail with reference to FIG. 6.

[0053] In the present disclosure, the artificial intelligence model (230) may be a single artificial intelligence model or multiple artificial intelligence models. The artificial intelligence model may be composed of a neural network (or artificial neural network) and may include statistical learning algorithms that mimic biological neurons in machine learning and cognitive science. A neural network may refer to a general model having a problem-solving function by changing the binding strength of synapses through learning, in which artificial neurons (nodes) that form a network by combining synapses. The neurons of the neural network may include a combination of weights or biases. The neural network may include one or more layers composed of one or more neurons or nodes. For example, the neural network may include an input layer, a hidden layer, and an output layer. The neural network can infer a desired result (output) from an arbitrary input (input) by changing the weights of neurons through learning.

[0054] The electronic device (200) can create a neural network, train or learn a neural network, perform a calculation based on received input data, generate an information signal based on the performance result, or retrain a neural network. The models of the neural network may include, but are not limited to, various types of models such as CNN (Convolution Neural Network) such as GoogleNet, AlexNet, VGG Network, R-CNN (Region with Convolution Neural Network), RPN (Region Proposal Network), RNN (Recurrent Neural Network), S-DNN (Stacking-based deep Neural Network), S-SDNN (State-Space Dynamic Neural Network), Deconvolution Network, DBN (Deep Belief Network), RBM (Restrcted Boltzman Machine), Fully Convolutional Network, LSTM (Long Short-Term Memory) Network, Classification Network, etc. The electronic device (200) may include one or more processors (e.g., 220 of FIG. 1) for performing operations according to models of a neural network.

[0055] Neural networks include CNN (Convolutional Neural Network), RNN (Recurrent Neural Network), perceptron, multilayer perceptron, FF (Feed Forward), RBF (Radial Basis Network), DFF (Deep Feed Forward), LSTM (Long Short Term Memory), GRU (Gated Recurrent Unit), AE (Auto Encoder), VAE (Variational Auto) Encoder), DAE (Denoising Auto Encoder), SAE (Sparse Auto Encoder), MC (Markov Chain), HN (Hopfield Network), BM (Boltzmann Machine), RBM (Restricted Boltzmann Machine), DBN (Depp Belief Network), DCN (Deep Convolutional Network), DN (Deconvolutional Network), DCIGN (Deep Convolutional Inverse Graphics Network), Generative Adversarial Network (GAN), Liquid State Machine (LSM), Extreme Learning Machine (ELM), It will be understood by those skilled in the art that any neural network may be included, including but not limited to ESN (Echo State Network), DRN (Deep Residual Network), DNC (Differentiable Neural Computer), NTM (Neural Turning Machine), CN (Capsule Network), KN (Kohonen Network), and AN (Attention Network).

[0056] According to an exemplary embodiment of the present disclosure, the electronic device (200) may be configured to include a CNN (Convolution Neural Network) such as GoogleNet, AlexNet, VGG Network, etc., a R-CNN (Region with Convolution Neural Network), an RPN (Region Proposal Network), an RNN (Recurrent Neural Network), an S-DNN (Stacking-based deep Neural Network), an S-SDNN (State-Space Dynamic Neural Network), a Deconvolution Network, a DBN (Deep Belief Network), an RBM (Restrcted Boltzman Machine), a Fully Convolutional Network, an LSTM (Long Short-Term Memory) Network, a Classification Network, Generative Modeling, eXplainable AI, Continual AI, Representation Learning, AI for Material Design, BERT for natural language processing, SP-BERT, MRC / QA, Text Analysis, Dialog System, GPT-3, GPT-4, GPT-5, Visual Analytics for vision processing, Visual Understanding, Video Synthesis, Anomaly Detection for ResNet data intelligence, Various artificial intelligence structures and algorithms can be used, including but not limited to Prediction, Time-Series Forecasting, Optimization, Recommendation, and Data Creation.

[0057] FIG. 3 is a flowchart of an operation method of a pet loss response system according to one embodiment.

[0058] In FIG. 3, steps S310 to S330 can be understood to be performed by the processor (220) of the electronic device (200).

[0059] In step S310, the electronic device (200) can receive GPS data at the time of the pet's loss and image data taken of the pet prior to the loss, via the first user terminal (100), which is the pet's guardian's terminal. This step can be triggered when the guardian, upon recognizing the pet's loss, enters information related to the pet via the first user terminal (100).

[0060] In step S320, the electronic device (200) can generate a digital leaflet (e.g., 501 of FIG. 5) based on GPS data and image data at the time of loss using the first artificial intelligence model (231). Step S320 will be described in detail with reference to FIGS. 4 and 5.

[0061] In step S330, the electronic device (200) may utilize the second artificial intelligence model (232) to determine a second user terminal (300) to which to send a notification including a digital flyer. Step S330 will be described in detail with reference to FIGS. 6 and 7.

[0062] FIG. 4 is a diagram for explaining a first artificial intelligence model according to one embodiment.

[0063] Referring to Fig. 4, the architecture of the first artificial intelligence model (231) can be confirmed. The first artificial intelligence model (231) can output real-time image data (441) of the pet based on GPS data (402) at the time of the pet's loss and image data (401) taken of the pet before the time of the loss.

[0064] The first artificial intelligence model (231) may include an image encoder (410), a time embedding module (420), a fusion module (430), and an image decoder (440).

[0065] An image encoder (410) may output a feature map (411) of the external features of a companion animal based on image data (401). The image encoder (410) may be a CNN-based layer. The feature map (411) output by the image encoder (410) may include information about the visual features of the companion animal (e.g., fur length, color, facial features, species).

[0066] The time embedding module (420) can embed the time difference between the GPS data (402) and the image data (401) at the time of loss. The time embedding module (420) can output a time difference embedding (421). The time embedding module (420) can calculate the time difference based on the timestamp of the GPS data (402) and the timestamp of the image data (401) and convert it into an embedding vector. The time difference embedding (421) can be combined with the feature map (411) in the fusion module (430) and used to conditionally reflect changes according to the time difference (e.g., changes in hair length and facial features). The time embedding module (420) can be implemented through a plurality of dense layers.

[0067] The fusion module (430) can combine the feature map (411) and the time difference embedding (421) to output a fused feature map (not shown). The fusion module (430) can generate γ parameters and β parameters of the same dimension as the feature map (411) based on the time difference embedding (421), and can apply scale adjustment or shift adjustment for each channel of the feature map (411) based on the γ parameters and β parameters. The fusion module (430) can be implemented based on FiLM (Feature-wise Linear Modulation).

[0068] The image decoder (440) can output real-time image data (441) of the companion animal based on a fused feature map (not shown). The image decoder (440) can be implemented with an upsampling layer, a residual block, and / or an attention block.

[0069] The electronic device (200) can generate a digital flyer containing real-time image data (441) of the companion animal.

[0070] FIG. 5 is an example of a digital flyer generated using a first artificial intelligence model according to one embodiment.

[0071] Referring to Fig. 5, a digital flyer (501) can be viewed. The digital flyer (501) may include various information for finding a companion animal.

[0072] The digital flyer (501) may include real-time image data (511) of the companion animal. The real-time image data (511) may be generated by the first artificial intelligence model (231) as described above (e.g., 441 of FIG. 4). However, the real-time image data (511) included in the digital flyer (501) may have a filter applied that matches the surrounding environment. For example, if the companion animal is in a muddy area, a filter that adjusts the hue or saturation to match the environment may be applied. That is, the electronic device (200) may apply a filter that matches the surrounding environment to the real-time image data (441) based on the companion animal's real-time location data, and then include the image in the digital flyer (501) (511).

[0073] The basic information section (512) of the digital flyer (501) may include the pet's name, breed, age, gender, neutering status, and characteristics. This allows for the identification of the lost pet's physical characteristics and basic personal information. The information included in the basic information section may have been entered by the pet's owner.

[0074] The Lost Information section (513) of the digital flyer (501) may include the date, time, and exact location of the lost pet. This information can help anyone who witnessed or cares for the pet quickly recognize and provide assistance. Additionally, additional descriptions can provide information about the circumstances and surroundings of the lost pet, helping witnesses understand the situation. The information included in the Lost Information section may be entered by the pet's owner or based on GPS data at the time of the pet's loss.

[0075] FIG. 6 is a diagram for explaining a second artificial intelligence model according to one embodiment.

[0076] Referring to FIG. 6, the architecture of the second artificial intelligence model (232) can be confirmed. The second artificial intelligence model (232) can utilize GPS data (402) at the time of the pet's loss, the pet's behavioral pattern information (601), and surrounding environment information (602). The second artificial intelligence model (232) can utilize the above data and / or information to determine the second user terminal (300) to which to send a notification including a digital flyer (501).

[0077] As described above, the companion animal wearable device (101) may include a short-range communication module (e.g., Bluetooth, Bluetooth low energy, ZigBee, WFD (WiFi direct), UWB (ultra-wide band), infrared data association (IrDA), near field communication (NFC)) instead of a mobile communication module. In this case, if the companion animal wearable device (101) goes out of the signal range with the first user terminal (100), it is difficult to obtain the companion animal's location data.

[0078] Accordingly, the second artificial intelligence model (232) may be implemented to first infer real-time location data (621) of the pet, and then adaptively determine the second user terminal (300) to which to send a notification by considering the real-time location data (621) and the elapsed time since the pet was lost (612) (e.g., 631).

[0079] The second artificial intelligence model (232) may include a preprocessing module (610), a simulation-based location inference module (620), and a notification area determination module (630).

[0080] The preprocessing module (610) can output initial location data (611), elapsed time since loss (612), and condition vector (613) based on GPS data (402) at the time of loss, the pet's behavioral pattern information (601), and surrounding environment information (602). The pet's behavioral pattern information (601) may be processed data collected from the pet wearable device (101). The surrounding environment information (602) may be information about the surrounding environment where the pet was lost (e.g., information about a city, rural area, riverside, muddy location, etc.).

[0081] The preprocessing module (610) can output initial location data (611) when a pet is lost based on GPS data (402). The preprocessing module (610) can output the elapsed time since loss (612) by comparing the current time with the time of loss (e.g., a timestamp included in the GPS data (402)). The preprocessing module (610) can output a condition vector (613) that combines (e.g., concatenates) the pet's behavioral pattern information (601) and surrounding environment information (602).

[0082] The simulation-based location inference module (620) can output real-time location data (621) of the pet based on initial location data (611), elapsed time since loss (612), and condition vector (613).

[0083] The simulation-based location inference module (620) may be based on a dynamic prediction network and rule-based simulation. The dynamic prediction network may include an RNN, an LSTM, and / or a feedforward network, and may output a predicted location of the companion animal. The rule-based simulation may include a random walk reflecting a certain level of randomness, a destination directionality reflecting the likelihood of the companion animal moving to a specific destination, and obstacle avoidance rules that avoid physical constraints of the surrounding environment. That is, the companion animal's real-time location data (621) may correspond to a weighted result of the predicted location of the companion animal derived from the dynamic prediction network and the correction value generated from the rule-based simulation.

[0084] In the rule-based simulation phase, at each time step, the simulation agent can determine virtual movements based on three key behavioral rules. First, random walks allow the pet to move randomly within the environment, generating diverse movement paths that reflect predictive uncertainty. Second, destination orientations can generate movement paths that reflect the pet's tendency to move toward specific destinations (e.g., home, food sources, etc.) or safety zones. Third, obstacle avoidance rules can refer to the process of avoiding obstacles and selecting the optimal path based on map information or environmental data. All of these processes can be combined with the pet's predicted location (or pattern) derived from the dynamic prediction network to improve prediction accuracy.

[0085] The notification area determination module (630) can output a notification area and a notification transmission target (631) based on real-time location data (621) and the elapsed time since loss (612). This will be described in more detail with reference to FIG. 7.

[0086] FIG. 7 is a diagram illustrating a notification area and a notification transmission target in which a notification including a digital flyer is provided according to one embodiment.

[0087] Referring to FIG. 7, the notification area determination module (630) can determine the area and target to which a notification will be sent based on the elapsed time since the pet was lost. The notification area determination module (630) can set an area included within a specific radius based on the real-time location data of the pet (e.g., the pet's location at the time of loss (701) -> the pet's location (702) updated (inferred) over time), and determine a user group located within the area as the target to which the notification will be sent.

[0088] The notification area determination module (630) may set a first area within a first radius centered on the real-time location data (701) of the pet as a notification area if the elapsed time since the pet was lost is less than a first threshold. At this time, users included within the first area are classified into a first recipient group, and notifications are sent to them with priority.

[0089] If the elapsed time since the loss is greater than or equal to the first threshold and less than the second threshold, the notification area determination module (630) may set a second area as the notification area by applying a wider radius centered on the pet's real-time location data (702). Users included within the second area are classified into a second recipient group, and notifications may also be sent to them.

[0090] At this time, the first threshold is smaller than the second threshold, and the first radius is also smaller than the second radius. Therefore, the types of notification transmission targets included in the first recipient group are smaller than or equal to the types of notification transmission targets included in the second recipient group. For example, the first recipient group may be composed of user terminals on which the application of the business operator operating the system (10) is installed and terminals of an animal hospital affiliated with the business operator operating the system (10), and the second recipient group may be composed of terminals of an animal protection organization affiliated with the business operator operating the system (10) in addition to the first recipient group.

[0091] That is, as time passes, the notification area and types of notification targets expand, so notifications can be sent to more and more users.

[0092] The notification area determination module (630) operates in such a way that it quickly identifies the location of a lost pet, initially sends notifications only to nearby users, and then expands the notification range over time to enable more effective searching.

[0093] The collection device (e.g., a user terminal or electronic device) according to the embodiments disclosed in this document may take various forms. The collection device may include, for example, a portable communication device (e.g., a smartphone), a computer device, a portable multimedia device, a portable medical device, a camera, a wearable device, or a home appliance. The collection device according to the embodiments of this document is not limited to the aforementioned devices.

[0094] The embodiments of this document and the terminology used herein are not intended to limit the technical features described in this document to specific embodiments, but should be understood to include various modifications, equivalents, or substitutes of the embodiments. In connection with the description of the drawings, similar reference numerals may be used for similar or related components. The singular form of a noun corresponding to an item may include one or more of the items, unless the context clearly indicates otherwise. In this document, each of the phrases "A or B", "at least one of A and B", "at least one of A or B", "A, B, or C", "at least one of A, B, and C", and "at least one of A, B, or C" can include any one of the items listed together in the corresponding phrase among those phrases, or all possible combinations thereof. Terms such as "first," "second," or "first" or "second" may be used merely to distinguish one component from another, and do not limit the components in any other respect (e.g., importance or order). When a component (e.g., a first component) is referred to as "coupled" or "connected" to another component (e.g., a second component), with or without the terms "functionally" or "communicatively," it means that the component can be connected to the other component directly (e.g., wired), wirelessly, or through a third component.

[0095] The term "module" used in the embodiments of this document may include a unit implemented in hardware, software, or firmware, and may be used interchangeably with terms such as logic, logic block, component, or circuit. A module may be an integral component, or a minimum unit or part of such a component that performs one or more functions. For example, according to one embodiment, a module may be implemented in the form of an application-specific integrated circuit (ASIC).

[0096] Embodiments of the present document may be implemented as software (e.g., a program) including one or more instructions stored in a storage medium (e.g., built-in memory or external memory) readable by a machine (e.g., an electronic device). For example, a processor (e.g., a processor) of the machine (e.g., an electronic device) may call at least one instruction among the one or more instructions stored from the storage medium and execute it. This enables the machine to operate to perform at least one function according to the at least one instruction called. The one or more instructions may include code generated by a compiler or code executable by an interpreter. The machine-readable storage medium may be provided in the form of a non-transitory storage medium. Here, 'non-transitory' only means that the storage medium is a tangible device and does not contain a signal (e.g., electromagnetic waves), and this term does not distinguish between cases where data is stored semi-permanently and cases where it is stored temporarily in the storage medium.

[0097] According to one embodiment, the method according to one embodiment disclosed in the present document may be provided as included in a computer program product. The computer program product may be traded as a product between a seller and a buyer. The computer program product may be distributed in the form of a machine-readable storage medium (e.g., compact disc read-only memory (CD-ROM)), or may be distributed online (e.g., downloaded or uploaded) via an application store (e.g., Play Store™) or directly between two user devices (e.g., smart phones). In the case of online distribution, at least a portion of the computer program product may be temporarily stored or temporarily generated in a machine-readable storage medium, such as the memory of a manufacturer's server, an application store's server, or an intermediary server.

[0098] According to one embodiment, each component (e.g., a module or a program) of the above-described components may include one or more entities, and some of the entities may be separated and placed in other components. According to one embodiment, one or more components or operations of the aforementioned components may be omitted, or one or more other components or operations may be added. Alternatively or additionally, a plurality of components (e.g., a module or a program) may be integrated into a single component. In this case, the integrated component may perform one or more functions of each of the plurality of components identically or similarly to those performed by the corresponding component among the plurality of components prior to the integration. According to one embodiment, the operations performed by a module, program, or other component may be executed sequentially, in parallel, iteratively, or heuristically, or one or more of the operations may be executed in a different order, omitted, or one or more other operations may be added.

Claims

1. In the pet loss response system, A pet wearable device comprising a GPS module that is worn by a pet and generates GPS data of the pet; A first user terminal, as a terminal of the guardian of the companion animal, which receives the GPS data from the companion animal wearable device; An electronic device that generates a digital flyer in the event of loss of said pet; and A second user terminal that receives a notification including the digital leaflet from the electronic device, The above electronic device, Through the first user terminal, GPS data at the time of loss of the pet and image data taken of the pet before the time of loss are received, Using the first artificial intelligence model, the digital leaflet is generated based on the GPS data and the image data at the time of loss, A system that uses a second artificial intelligence model to determine the second user terminal to which a notification including the digital flyer will be sent.

2. In paragraph 1, The above first artificial intelligence model is, A system characterized in that it compares a timestamp included in the GPS data at the time of loss with a timestamp included in the image data, and generates real-time image data of the pet based on the image data.

3. In paragraph 2, The above first artificial intelligence model is, An image encoder that outputs a feature map of the external features of the companion animal based on the image data; A time embedding module that embeds the time difference between the two data based on the GPS data and the image data at the time of loss; A fusion module that combines the above feature map and time difference embedding to output a fused feature map; and An image decoder that outputs real-time image data of the companion animal based on the fused feature map is included. The above electronic device, A system for generating a digital flyer including real-time image data of the companion animal.

4. In paragraph 3, The above fusion module, A system characterized in that it generates γ parameters and β parameters of the same dimension as the feature map based on the above time difference embedding, and applies scale adjustment or shift adjustment for each channel of the feature map based on the γ parameters and the β parameters.

5. In paragraph 1, The above second artificial intelligence model is, A system characterized in that it infers real-time location data of the pet and adaptively determines the second user terminal to which the notification will be sent by considering the real-time location data and the elapsed time since the pet was lost.

6. In paragraph 5, The above second artificial intelligence model is, A preprocessing module that outputs initial location data, elapsed time since the loss, and a condition vector based on GPS data at the time of loss, behavioral pattern information of the pet, and surrounding environment information; A simulation-based location inference module that outputs real-time location data of the pet based on the initial location data, the elapsed time since the loss, and the condition vector; and A system comprising a notification area determination module that outputs a notification area and a notification transmission target based on the real-time location data and the elapsed time since the loss.

7. In paragraph 6, The real-time location data of the above pet is: The predicted location of the above companion animal derived from the dynamic prediction network and the correction value derived from the rule-based simulation are weighted and combined. The above rule-based simulation is, A system characterized by a random walk that reflects a certain level of randomness, a destination directionality that reflects the likelihood that the companion animal will move to a specific destination, and an obstacle avoidance rule that avoids physical constraints of the surrounding environment.

8. In paragraph 6, The above notification area determination module, If the elapsed time since the loss is less than the first threshold, a first area included within a first radius based on the real-time location data of the pet is determined as a notification area, and a first group of recipients located within the first area is determined as a notification transmission target. If the elapsed time since the loss is less than the second threshold, a second area included within a second radius based on the real-time location data of the pet is determined as a notification area, and a second group of recipients located within the second area is determined as a notification transmission target. The above first threshold is smaller than the above second threshold, The first radius is smaller than the second radius, A system characterized in that the type of notification transmission target included in the first recipient group is less than or equal to the type of notification transmission target included in the second recipient group.

9. In paragraph 6, The above first artificial intelligence model is, A system characterized in that it generates the digital flyer by applying a filter suitable for the surrounding environment based on the real-time location data of the companion animal.

10. In the method of operating an electronic device, A step of receiving GPS data at the time of loss of the pet and image data taken of the pet before the time of loss through a first user terminal, which is a terminal of the pet's guardian; A step of generating the digital leaflet based on the GPS data and the image data at the time of loss using the first artificial intelligence model; and A step of determining a second user terminal to which a notification including the digital flyer will be sent by utilizing a second artificial intelligence model, The GPS data at the time of loss of the above pet is: A method generated from a pet wearable device worn by the pet and including a short-range communication module.

Images