Human-unidentifiable information collection device for driving hardware orchestration for interactive artificial intelligence learning

Abstract

A human-unidentifiable information collection device for artificial intelligence learning, according to one embodiment, may comprise: at least one microphone capturing sounds generated around a companion animal to generate first audio data and second audio data; an inertial measurement unit (IMU) for generating inertial data regarding a change in acceleration and angular velocity according to the movement of the companion animal; and a processor for adaptively determining respective sampling rates for collecting the first audio data, the second audio data, and the inertial data, on the basis of at least one of the breed, age, gender, neutering status, or temperament type of the companion animal.

Description

A human-anonymous data collection device that drives hardware orchestration for interactive artificial intelligence learning.

[0001] The following disclosure relates to a collection device for artificial intelligence learning, and more specifically, to a human-anonymous information collection device that drives hardware orchestration for artificial intelligence learning and a method of operating the same.

[0002] With the rise of nuclear families, single-person households, and an aging population, a growing number of modern people are experiencing loneliness. As a result, more and more people are recognizing their pets as family members. Pets enjoy daily life with their owners, and interest in their ability to express themselves is also on the rise.

[0003] As social perceptions of companion animals change, research is actively underway on various devices and services for people with companion animals.

[0004] A method for remotely managing a pet through a wearable device attached to the pet is proposed, and a device for collecting health information of the pet using a detection sensor mounted on the wearable device and measuring location information of the pet using a GPS module or beacon module mounted on the wearable device is also provided.

[0005] A device for collecting non-human identifiable information for artificial intelligence learning according to one embodiment may include at least one microphone for capturing sounds occurring around a companion animal to generate first audio data and second audio data; an inertial measurement unit (IMU) for generating inertial data on changes in acceleration and angular velocity according to movement of the companion animal; and a processor for adaptively determining respective sampling rates for collecting the first audio data, the second audio data, and the inertial data based on at least one of the breed, age, sex, neutering status, or temperament type of the companion animal.

[0006] According to one embodiment, the processor may define each interpolation model based on a multiple linear regression analysis algorithm utilizing the sampling rates of each of the first audio data, the second audio data, and the inertial data. The processor may interpolate the first audio data, the second audio data, and the inertial data based on each of the interpolation models.

[0007] According to one embodiment, the processor can manage an operating mode of the at least one microphone, the operating mode including an active mode and a low-power mode. The low-power mode may be a mode in which a sampling rate for collecting data is lower and the complexity of the interpolation model is lower compared to the active mode.

[0008] In one embodiment, the processor may set the at least one microphone to the activation mode when the second audio data identifies the voice of the pet parent of the pet or the inertial data identifies the walking state of the pet. The processor may set the at least one microphone to the low power mode when the inertial data identifies the sleeping state of the pet.

[0009] In one embodiment, the at least one microphone may include a first microphone including a filter that captures sound in an audible frequency band of the companion animal and outputs the first audio data in an inaudible frequency band of a human; and a second microphone that outputs the second audio data in an audible frequency band of a human.

[0010] According to one embodiment, the at least one microphone may be a microphone including a first filter that captures sound in an audible frequency band of the companion animal and outputs the first audio data in an inaudible frequency band of a human, and a second filter that outputs the second audio data in an audible frequency band of a human.

[0011] According to one embodiment, the collection device may further include a gas sensor that detects gases contained in the air around the pet to generate olfactory data. The olfactory data may have three concentration levels for each of familiar odors, unfamiliar odors, odors classified by race, odors of people the pet has met, odors of spaces the pet has visited, odors characteristic of household fabrics, odors of food frequently eaten, and odors according to the gender, race, cleanliness, food consumed, and health of the primary caregiver.

[0012] According to one embodiment, the processor may perform noise filtering on the first audio data, the second audio data, the inertial data, and the olfactory data. The processor may calibrate the first audio data, the second audio data, the inertial data, and the olfactory data together, on which the noise filtering has been performed.

[0013] In one embodiment, the collection device may further include a biometric sensor for measuring at least one of an electrocardiogram (ECG), a photoplethysmogram (PPG), or an electroencephalography (EEG) of the companion animal; a global positioning system (GPS) for measuring the location of the companion animal; and a camera for photographing at least a portion of the companion animal.

[0014] In one embodiment, the collection device may be implemented as a smart collar, a smart harness, a wearable device, or an accessory for the companion animal.

[0015] The collection device according to the present disclosure can effectively integrate and manage data acquired from different types of sensors.

[0016] Figure 1 illustrates a system architecture that performs artificial intelligence learning based on human-anonymous information according to one embodiment.

[0017] FIG. 2 illustrates a block diagram of a human-non-identifiable information collection device according to one embodiment.

[0018] FIG. 3 is a diagram illustrating data output by a sensor of a human-non-identifiable information collection device according to one embodiment.

[0019] FIG. 4 is a diagram for explaining a sampling rate of a sensor according to one embodiment.

[0020] FIG. 5 is a drawing for explaining an operation mode of a sensor according to another embodiment.

[0021] FIG. 6 is a diagram illustrating sensor orchestration according to one embodiment.

[0022] FIG. 7 illustrates an example hardware implementation of a human-non-identifiable information collection device according to one embodiment.

[0023] 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.

[0024] 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.

[0025] 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.

[0026] 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.

[0027] 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.

[0028] 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).

[0029] 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.

[0030] 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.

[0031] Figure 1 illustrates a system architecture that performs artificial intelligence learning based on human-anonymous information according to one embodiment.

[0032] Referring to FIG. 1, a system (10) according to one embodiment can perform artificial intelligence learning based on human-non-identifiable information. The system (10) may include a human-non-identifiable information collection device (100) (hereinafter referred to as a "collection device") and a server (200).

[0033] In one embodiment, the collection device (100) can collect information (e.g., both human-identifiable information and non-human-identifiable information) surrounding the companion animal (11). The collection device (100) can be worn on the companion animal (11) and can be maintained at a predetermined distance from the companion animal (11). The information collected by the collection device (100) can be transferred to behavioral data (13) of the companion animal (11).

[0034] According to one embodiment, the collection device (100) and the server (200) (e.g., the server (108) of FIG. 1) may be connected via a network (12) (e.g., a local area network (LAN), a wide area network (WAN), a value added network (VAN), a mobile radio communication network, a satellite communication network, or a combination thereof). The collection device (100) and the server (200) may communicate with each other via a wired communication method or a wireless communication method (e.g., wireless LAN (WiFi), Bluetooth, Bluetooth low energy, ZigBee, WiFi direct (WFD), ultra wide band (UWB), infrared data association (IrDA), near field communication (NFC)).

[0035] According to one embodiment, the collection device (100) may include a sensor module (110), a processor (120), a memory (130), a communication module (140), and a power module (150). The sensor module (110) may detect information surrounding the companion animal (11). The sensor module (110) may include a plurality of sensors (or units) (e.g., see FIG. 2). The processor (120) (e.g., an application processor) may access the memory (130) and execute one or more instructions. The memory (130) may store various data used (or collected) by at least one component of the collection device (100) (e.g., the processor (120) or the sensor module (110)). The communication module (140) may support the establishment of a communication channel between the collection device (100) and the server (200) (or an external electronic device) and the performance of communication through the established communication channel. The power module (150) can supply power to at least one component of the collection device (100). The power module (150) can include a rechargeable secondary battery or fuel cell. When the power module (150) is connected to power and the collection device (100) is in a charged state, the collection device (100) can activate the communication module (140) to transmit the collected information to the server (200).

[0036] According to one embodiment, the server (200) can train an artificial intelligence model based on information collected by the collection device (100) (e.g., both human-identifiable information and human-non-identifiable information). The training of the artificial intelligence model can be performed by the artificial intelligence module (210), and the server (200) can utilize an accelerator (220) and a processor (230) together when training the artificial intelligence model. The server (200) can store the information collected by the collection device (100) (e.g., both human-identifiable information and human-non-identifiable information) in a database (240) and utilize the same when training the artificial intelligence model.

[0037] According to one embodiment, the collection device (100) can efficiently collect human-identifiable information as well as human-non-identifiable information by simulating the sensory organs of an animal (e.g., a companion animal (11)). 1) The collection device (100) may include a microphone (e.g., at least one microphone (112) of FIG. 2) for collecting sounds in an inaudible frequency band for humans, and a gas sensor (e.g., a gas sensor (113) of FIG. 2) for collecting odors that cannot be detected by humans. 2) Companion animals may also have different sensitive and insensitive sensory organs depending on the breed, age, sex, neutering status, temperament, etc. Considering the different sensitivities of each sensory organ, the collection device (100) may adaptively determine the sampling rate of the sensor corresponding to the sensory organ. For example, the sampling rate of the sensor corresponding to the sensitive sensory organ may be determined to a high value. 3) Depending on the pet's condition (e.g., during a walk, while sleeping, or immediately after recognizing the voice of its primary caregiver), the sensitivity of its sensory organs may vary. When the collection device (100) identifies a specific condition of the pet, it can set the operating mode of the sensors to an active mode or a low-power mode. By setting the operating mode of the sensors according to the pet's condition, the collection device (100) can be implemented as a low-power wearable device.

[0038] According to one embodiment, the collection device (100) may be designed with sensor orchestration in mind. Since the collection device (100) includes sensors (or units) corresponding to the animal's sensory organs, the collection device (100) may be a multi-sensor device. The collection device (100) can effectively integrate and manage data acquired from different types of sensors. For example, the data collected from each sensor (or unit) is a sampled result and thus needs to be interpolated. In addition, as described above, since the collection device (100) includes multiple sensors (or units) with different sampling rates, it is necessary to comprehensively consider the characteristics of other data during data interpolation. The collection device (100) can define an interpolation model for each data based on a multiple linear regression algorithm utilizing each sampling rate. The structure and operation of the collection device (100) will be described in more detail below.

[0039] FIG. 2 is a block diagram of a human-non-identifiable information collection device according to one embodiment, and FIG. 3 is a diagram for explaining data output by a sensor of a human-non-identifiable information collection device according to one embodiment.

[0040] Referring to FIG. 2, according to one embodiment, the collection device (100) can collect human-non-identifiable information (and / or human-identifiable information) for artificial intelligence learning through a sensor module (110). The sensor module (110) can include an inertial measurement device (111), at least one microphone (112), a gas sensor (113), a biometric sensor (114), a camera (115), and a global positioning system (GPS) (116).

[0041] According to one embodiment, the inertial measurement device (111) may generate inertial data. Referring to FIG. 3, the inertial data (311) may be time-series data, which may be data on changes in acceleration and angular velocity according to the movement of the companion animal. The inertial measurement device (111) may include an acceleration sensor and a gyro sensor, and may be referred to as an inertial sensor.

[0042] According to one embodiment, at least one microphone (112) can capture sounds occurring around the pet to generate first audio data (312) and second audio data (313). The at least one microphone (112) can include a first microphone (112-1) (e.g., an all band MIC) that captures sounds in an audible frequency band of the pet. The first microphone (112-1) can include a filter (e.g., a band pass filter or a high pass filter) that outputs first audio data (312) in an inaudible frequency band for humans (e.g., audio data in a band of 20 to 40 kHz). The at least one microphone (112) can include a second microphone (112-2) (e.g., a stereo LFA MIC) that captures sounds in an audible frequency band for humans. The second microphone (112-2) can output second audio data (313) (e.g., audio data in a band of 0 to 20 kHz).

[0043] In one embodiment, at least one microphone (112) may include only one microphone that captures sounds in the audible frequency band of the companion animal. In this case, the one microphone may include a first filter (e.g., a band-pass filter or a high-pass filter) that outputs first audio data (312) in the inaudible frequency band of a human, and a second filter (e.g., a low-pass filter) that outputs second audio data (313) in the audible frequency band of a human.

[0044] In one embodiment, the third audio data (314) may correspond to the voice of the primary caregiver of the companion animal. The third audio data (314) may be collected from an external electronic device of the collection device (100), and the third audio data (3114) may be received via the communication module (140).

[0045] According to one embodiment, the gas sensor (113) can detect gases contained in the air around the pet to generate olfactory data (315). The olfactory data (315) can have three concentration levels for each of a familiar smell to the pet, an unfamiliar smell, a smell according to race classification, a smell of a person the pet has met, a smell of a space the pet has been to, a smell of fabric unique to the house, a smell of food mainly eaten, and a smell according to the gender, race, cleanliness, food consumed, and health of the primary caretaker.

[0046] In one embodiment, environmental data (315) and profile data (317) may be collected from an external electronic device (or server) of the collection device (100). Environmental data (316) may include information about the temperature, humidity, and / or location of the environment in which the pet is located. Profile data (317) may include a profile of the pet (e.g., breed, age, gender, neutering status, and / or temperament type).

[0047] According to one embodiment, the biometric sensor (114) can generate biometric data (318). The biometric data (318) can include an electrocardiogram (ECG), a photoplethysmogram (PPG), and / or an electroencephalography (EEG) of the companion animal.

[0048] In one embodiment, the camera (115) may generate video data (319) (or image data). The video data (319) may be a photograph of at least a portion of the pet (e.g., a tail). The Global Positioning System (GPS) (116) may be used to measure the location of the pet. Based on the GPS (116), the collection device (100) may receive environmental data (316) regarding the environment in which the pet is located.

[0049] Referring to FIG. 2, according to one embodiment, the processor (120) may perform sampling rate determination, operation mode determination, data interpolation, noise filtering, and / or calibration. The specific operation of the processor (120) will be described in detail with reference to FIGS. 4 to 6.

[0050] FIG. 4 is a diagram for explaining a sampling rate of a sensor according to one embodiment.

[0051] Referring to FIG. 4, it can be seen that the data output by the sensor has different formats (411, 412, and 413) depending on the sampling rate of the sensor. The sampling rate of the sensor refers to the speed at which the sensor samples information, and may correspond to the frequency at which the sensor collects information. The sampling rate of the sensor may be expressed as the number of data output by the sensor per second.

[0052] For example, if a gas sensor measures gas once per second, its sampling rate might be 1 Hz. Sensors such as cameras can have higher sampling rates than gas sensors, meaning they can measure the environment with higher resolution and greater accuracy than gas sensors.

[0053] A higher sampling rate allows a sensor to detect environmental changes more quickly, but the larger the amount of information collected, the more difficult it may be to process. Conversely, a sampling rate that is too low may not be able to properly detect environmental changes. Therefore, the sensor sampling rate needs to be set differently depending on the sensor's requirements.

[0054] Companion animals may differ in breed, age, sex, neutering status, temperament, etc., and may have different sensitive and insensitive sense organs for each individual. A collection device according to one embodiment (e.g., collection device (100) of FIG. 1) may adaptively determine a sampling rate of a sensor by considering different sensitivities for each sense organ. For example, the sampling rate of a sensor corresponding to a sensitive sense organ may be determined to be a high value. That is, the collection device (100) may adaptively determine respective sampling rates for collecting first audio data, second audio data, inertial data, and olfactory data after obtaining profile data (e.g., profile data (317) of FIG. 3) of a companion animal (e.g., companion animal (11) of FIG. 1) through a communication module (e.g., communication module (140) of FIG. 1).

[0055] According to one embodiment, the collection device (100) can efficiently collect human-identifiable information as well as human-non-identifiable information by imitating the sensory organs of an animal.

[0056] FIG. 5 is a drawing for explaining an operation mode of a sensor according to another embodiment.

[0057] To enhance the practicality of wearable devices, low-power implementation can be an important issue. A collection device (e.g., collection device (100) of FIG. 1) according to one embodiment can manage the operating modes of sensors (e.g., 111 to 116 of FIG. 2) included in a sensor module (e.g., sensor module (110) of FIG. 1). The operating modes can include an active mode and a low-power mode. The low-power mode can be a mode in which the sampling rate for collecting data is lower than that of the active mode.

[0058] Referring to FIG. 5, for example, the collection device (100) may set at least one microphone (112), a gas sensor (113), and a camera (115) to a low power mode. When the collection device (100) identifies the sleeping state of the companion animal through inertial data (e.g., inertial data (311) of FIG. 3), the collection device (100) may set at least one microphone (112), a gas sensor (113), and / or a camera (115) to a low power mode.

[0059] According to one embodiment, the collection device (100) may set at least one microphone (112), gas sensor (113), and / or camera (115) to an active mode when the voice of the primary caregiver of the companion animal is identified through second audio data (e.g., second audio data (314) of FIG. 3) or when the walking state of the companion animal is identified through inertial data (313).

[0060] Depending on the pet's state (e.g., during a walk, while sleeping, or immediately after recognizing the voice of its primary caregiver), the sensitivity of its sensory organs may vary. By including a function for identifying a specific state of the pet and setting the operating mode of the sensors based on the identified specific state, the collection device (100) according to one embodiment can be implemented as a low-power wearable device.

[0061] As mentioned above, low-power mode allows for a lower sampling rate for data collection compared to active mode, as well as a lower complexity for the interpolation model. Interpolation will be described below.

[0062] FIG. 6 is a diagram illustrating sensor orchestration according to one embodiment.

[0063] Referring to FIG. 6, since the data output from each sensor (or unit) included in the sensor module (e.g., the sensor module (110) of FIG. 1) is a sampled result (e.g., 601), the collection device (e.g., the collection device (100) of FIG. 1) needs to interpolate the data. However, depending on how the interpolation model is defined, the shape of the interpolated data may be different (e.g., see 611 and 612). When utilizing the polynomial interpolation method, accuracy and complexity have a trade-off relationship in the interpolation model. Referring to mathematical expression 1, the polynomial interpolation model can be confirmed.

[0064] [Mathematical Formula 1]

[0065]

[0066] In mathematical expression 1, y is a polynomial interpolation model, n-1 is the degree of the polynomial interpolation model, and a n-1 can be a coefficient. Since the accuracy of the interpolated data and the complexity of the interpolation model may differ depending on the degree of the polynomial interpolation model, it is necessary to determine the appropriate degree of the polynomial interpolation model for each data.

[0067] In addition, as described above, since the sampling rate of each sensor is different, the collection device (100) needs to comprehensively consider the characteristics of other data when interpolating data.

[0068] According to one embodiment, the collection device (100) can define an interpolation model for each data (e.g., determine the degree of the interpolation model) based on a multiple linear regression algorithm that utilizes the sampling rate of each sensor. The multiple linear regression algorithm may be an algorithm that models a linear relationship between a dependent variable and one or more independent variables. Referring to Equation 2, the mathematical formula of the multiple linear regression algorithm can be confirmed.

[0069] [Equation 2]

[0070]

[0071] In Equation 2, Y is the dependent variable, X_N is the Nth independent variable, Beta_N is the weight multiplied by the Nth independent variable, and e can be an error.

[0072] According to one embodiment, the collection device (100) can utilize the sampling rates of each of the first audio data, the second audio data, the inertial data, and the olfactory data in a multi-linear algorithm. Specifically, by inputting the sampling rate of the data into each of the independent variables X_N, the degree of the polynomial interpolation model can be obtained as the dependent variable Y. At this time, the weights and errors multiplied by the independent variables can be experimentally obtained for each data (e.g., each sensor).

[0073] According to one embodiment, the collection device (100) can define a polynomial interpolation model by inputting sampling data since the degree of the polynomial interpolation model has been determined. Thereafter, the collection device (100) can interpolate each piece of data based on the polynomial interpolation model.

[0074] That is, the collection device (100) according to one embodiment may be designed with sensor orchestration in mind. Since the collection device (100) includes sensors (or units) corresponding to the animal's sensory organs, the collection device (100) may be a multi-sensor device. The collection device (100) can effectively integrate and manage data acquired from different types of sensors.

[0075]

[0076] FIG. 7 illustrates an example hardware implementation of a human-non-identifiable information collection device according to one embodiment.

[0077] Referring to FIG. 7, according to one embodiment, a collection device (e.g., the collection device (100) of FIG. 1) may be implemented in a form that can be worn by a companion animal. For example, the collection device (100) may be implemented as a smart collar (701), a smart harness (702), or a wearable device (703). However, the collection device (100) may also be implemented as an accessory that maintains a predetermined distance from the companion animal.

[0078]

[0079] The collection 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.

[0080] 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 (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.

[0081] 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).

[0082] 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.

[0083] 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.

[0084] 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 a device for collecting human-anonymous information for artificial intelligence learning, At least one microphone that captures sounds occurring around the pet and generates first audio data and second audio data; An inertial measurement unit (IMU) that generates inertial data on changes in acceleration and angular velocity according to the movement of the companion animal; and A collection device comprising a processor that determines respective sampling rates for collecting the first audio data, the second audio data, and the inertial data based on profile data including at least one of the breed, age, sex, neutering status, or temperament type of the companion animal, and defines respective interpolation models based on a multiple linear regression algorithm utilizing the respective sampling rates of the first audio data, the second audio data, and the inertial data.

2. In paragraph 1, The above processor, A collection device that interpolates the first audio data, the second audio data, and the inertial data based on each of the interpolation models.

3. In paragraph 2, The processor manages an operating mode including an activation mode and a low power mode of the at least one microphone, The above low power mode is a collection device in which the sampling rate for collecting data is lower and the complexity of the interpolation model is lower than that of the above active mode.

4. In paragraph 3, The above processor, When the voice of the primary caregiver of the companion animal is identified through the second audio data or the walking state of the companion animal is identified through the inertial data, the at least one microphone is set to the activation mode, A collection device that sets at least one microphone to the low power mode when the sleeping state of the companion animal is identified through the inertial data.

5. In paragraph 1, At least one microphone above, A first microphone including a filter that captures sounds in the audible frequency band of the companion animal and outputs the first audio data in the inaudible frequency band of a human; and A collection device comprising a second microphone for outputting the second audio data in the human audible frequency band.

6. In paragraph 1, At least one microphone above, A collection device, which is a microphone including a first filter that captures sounds in the audible frequency band of the companion animal and outputs the first audio data in the inaudible frequency band of a human, and a second filter that outputs the second audio data in the audible frequency band of a human.

7. In paragraph 1, Further comprising a gas sensor that detects gas contained in the air around the companion animal and generates olfactory data, The above olfactory data is collected by a device having three concentration levels.

8. In paragraph 7, The above processor, Each interpolation model is defined based on a multiple linear regression algorithm utilizing the sampling rates of each of the first audio data, the second audio data, the inertial data, and the olfactory data, A collection device that interpolates the first audio data, the second audio data, the inertial data, and the olfactory data based on each of the interpolation models.

9. In paragraph 1, A biosensor for measuring at least one of an electrocardiogram (ECG), a photoplethysmogram (PPG), or an electroencephalography (EEG) of the companion animal; GPS (Global Positioning System) for measuring the location of the above companion animal; and A collection device further comprising a camera for photographing at least a portion of said companion animal.

10. In paragraph 1, The above collection device is a collection device implemented as a smart collar, a smart harness, a wearable device, or an accessory for the companion animal.

11. A step of capturing sounds occurring around the pet using an audio device to generate first audio data and second audio data; A step of generating inertial data on changes in acceleration and angular velocity according to the movement of the companion animal using an inertial measurement device; A step of using a processor to determine a sampling rate for collecting the first audio data, the second audio data, and the inertial data, based on at least one of the breed, age, sex, neutering status, or temperament type of the companion animal; and A method for collecting human-anonymous information for artificial intelligence learning, comprising a step of defining each interpolation model based on a multiple linear regression algorithm utilizing the sampling rates of each of the first audio data, the second audio data, and the inertial data using the processor.

12. A computer program stored on a recording medium that executes the method of claim 11 in combination with hardware.

Images