Pet sign detection method, detection device and detection system

Abstract

The invention provides a pet vital sign detection method, detection equipment and a detection system. The detection method comprises the steps of judging whether a target pet is located at a preset position or not based on detection data from an inertial sensor; and if the target pet is located at the preset position, determining the vital signs of the target pet based on the detection data. According to the pet vital sign detection method, detection equipment and detection system provided by the invention, daily health detection can be carried out on the pet, the body abnormity of the pet can be found in time in the early stage of the state of an illness of the pet, and the damage to the body of the pet can be reduced.

Description

Technical Field

[0001] This manual relates to the field of pet care technology, and in particular to methods, equipment and systems for detecting pet vital signs. Background Technology

[0002] With economic and social development and the continuous improvement of people's living standards, the popularity of pets, represented by cats and dogs, in households is rapidly increasing. Consequently, people are paying more and more attention to their pets' health. Because pets cannot speak, communication between humans and pets is often difficult. People often only discover abnormalities in their pets' health when their condition has become quite serious. Therefore, routine health checks for pets have become particularly important.

[0003] Therefore, there is a need for a new method, equipment, and system for detecting pet vital signs, which can be used for routine health checks on pets. This would allow for the timely detection of abnormalities in the early stages of a pet's illness, reducing the damage the pet suffers.

[0004] The information in the background section is merely information known only to the inventor and does not imply that such information had entered the public domain before the date of this application, nor does it imply that it can be considered prior art in this disclosure. Summary of the Invention

[0005] This manual provides methods, equipment, and systems for detecting pet vital signs, and can solve problems existing in related technologies.

[0006] Firstly, this specification provides a method for detecting the vital signs of a pet. The method includes: determining whether the target pet is in a preset position based on detection data from an inertial sensor; and if the target pet is in the preset position, determining the target pet's vital signs based on the detection data.

[0007] In some embodiments, determining whether a target pet is in a preset position based on detection data from an inertial sensor includes: receiving detection data from an inertial sensor; obtaining a current signal sequence based on the detection data, wherein the current signal sequence includes the signal of the current frame and signals of multiple historical frames that are in time before the current frame; performing feature extraction on the current signal sequence to obtain feature parameters corresponding to the current frame; and inputting the feature parameters into a pre-trained position detection model to determine whether the target pet is in a preset position.

[0008] In some embodiments, before inputting the feature parameters into a pre-trained position detection model, the method includes: receiving training data from an inertial sensor; wherein the training data is measured by the inertial sensor when the target pet is in a preset position and when it leaves the preset position; performing feature extraction on the training data to obtain training feature parameters of the same type as the feature parameters; and training the position detection model using the training feature parameters.

[0009] In some embodiments, before determining the vital signs of the target pet based on the detection data, the method includes: determining whether the target pet is in a resting state based on the detection data. If the target pet is in a preset location and in a resting state, the vital signs are determined based on the detection data.

[0010] In some embodiments, determining whether a target pet is in a resting state based on detection data includes: receiving detection data from an inertial sensor; obtaining a current signal sequence based on the detection data, wherein the current signal sequence includes the signal of the current frame and signals of multiple historical frames that are time-series prior to the current frame; performing feature extraction on the current signal sequence to obtain feature parameters corresponding to the current frame; and inputting the feature parameters into a pre-trained activity detection model to determine whether the target pet is in a resting state.

[0011] In some embodiments, before inputting the feature parameters into a pre-trained activity detection model, the method includes: receiving training data from an inertial sensor; wherein the training data is measured by the inertial sensor when the target pet is in a preset position and when it leaves the preset position; extracting features from the training data to obtain training feature parameters of the same type as the feature parameters; and training the activity detection model using the training feature parameters.

[0012] In some embodiments, the detection data includes initial current data and multiple frames of initial historical data that precede the initial current data in time. Both the initial current data and each frame of initial historical data have multiple data dimensions. Obtaining the current signal sequence based on the detection data includes: calculating the average value of the initial historical data of a first preset number of frames adjacent to the initial current data in each data dimension; subtracting the average value from the initial current data in the corresponding data dimension to obtain the signal of the current frame; and then using the signal of the current frame and the signals of multiple historical frames to form the current signal sequence.

[0013] In some embodiments, the signal of the current frame and the signal of each historical frame have multiple data dimensions; feature extraction of the current signal sequence to obtain feature parameters corresponding to the current frame includes: feature extraction of the current signal sequence to obtain multiple signal features related to the current frame, and combining the multiple signal features into a feature vector.

[0014] In some embodiments, the multiple signal features include features of the current frame; performing feature extraction on the current signal sequence to obtain multiple signal features related to the current frame includes: performing feature extraction on the signal of the current frame to obtain features of the current frame; wherein, the features of the current frame include at least one of the mean, maximum value, minimum value, median or standard deviation of the signal of the current frame.

[0015] In some embodiments, feature extraction of the current signal sequence to obtain multiple signal features related to the current frame includes: calculating the absolute value of the difference between the signal of the current frame and the signal of a historical frame adjacent to the current frame in each data dimension to obtain a differential signal; and performing feature extraction on the differential signal to obtain features of the differential signal; wherein the multiple signal features include features of the differential signal, and the features of the differential signal include at least one of the mean, maximum value, minimum value, median or standard deviation of the differential signal.

[0016] In some embodiments, feature extraction of the current signal sequence to obtain multiple signal features related to the current frame includes: forming a first sub-signal sequence by combining the signal of the current frame with the historical frame signal of a second preset number of adjacent frames; and performing feature extraction on the first sub-signal sequence to obtain features of the first sub-signal sequence; wherein the multiple signal features include features of the first sub-signal sequence, and the features of the first sub-signal sequence include at least one of the standard deviation or covariance of the first sub-signal sequence in each data dimension.

[0017] In some embodiments, the multiple signal features include a confidence feature; feature extraction of the current signal sequence to obtain multiple signal features related to the current frame includes: forming a second sub-signal sequence by combining the signal of the current frame with the historical frame signal of a third preset number of adjacent frames; calculating the confidence of each data dimension based on the second sub-signal sequence; and using at least one of the maximum value, mean, or median of the confidence of each data dimension as a confidence feature.

[0018] In some embodiments, vital signs include respiratory parameters; calculating the reliability of each data dimension based on the second sub-signal sequence includes: smoothing the data of the second sub-signal sequence in each data dimension to obtain a smoothed data sequence corresponding to the data dimension; performing a fast Fourier transform on the smoothed data sequence to obtain the actual frequency domain signal; performing peak finding across the entire frequency domain signal to obtain the actual maximum peak value; finding peaks in a preset interval of the actual frequency domain signal to obtain the target maximum peak value and the target second maximum peak value; wherein, the preset interval corresponds to the respiratory parameters; comparing the actual maximum peak value with the target maximum peak value; if the actual maximum peak value is equal to the target maximum peak value, then the reliability of the corresponding data dimension is equal to the ratio of the target maximum peak value to the target second maximum peak value; if the actual maximum peak value is not equal to the target maximum peak value, then the reliability of the corresponding data dimension is equal to the ratio of the actual maximum peak value to the target maximum peak value.

[0019] In some embodiments, vital signs include respiratory parameters; the current signal sequence has multiple data dimensions; determining the vital signs of the target pet based on the detection data includes: within a first preset time window, obtaining the signal of the data dimension with the highest confidence in the current signal sequence to form a respiratory signal sequence; performing bandpass filtering on the respiratory signal sequence in a first preset frequency band to obtain the respiratory waveform signal of the target pet; wherein the first preset frequency band is the respiratory frequency band of the pet; and determining respiratory parameters based on the respiratory waveform signal; wherein the respiratory parameters include at least one of respiratory rate, respiratory rhythm, or respiratory pattern.

[0020] In some embodiments, vital signs include heart rate parameters; determining the vital signs of the target pet based on detection data further includes: differentiating the respiratory signal sequence from the respiratory waveform signal to obtain a heart rate signal sequence; performing bandpass filtering on the heart rate signal sequence in a second preset frequency band to obtain the heart rate waveform signal of the target pet, wherein the second preset frequency band is the heart rate frequency band of the pet; and determining the heart rate parameters based on the heart rate waveform signal; wherein the heart rate parameters include at least one of heart rate or heart rate variability.

[0021] Secondly, this specification provides a device for detecting the vital signs of a pet. The device includes a main body, an inertial sensor, and a detection component. The main body includes a preset position for accommodating a target pet. The inertial sensor is disposed on the main body. The detection component is communicatively connected to the inertial sensor. The detection component includes at least one storage medium and at least one processor. The at least one storage medium stores at least one set of instructions for detecting the vital signs of the target pet. The at least one processor is communicatively connected to the at least one storage medium. When the detection device is running, the at least one processor reads the at least one set of instructions and executes the detection method according to the instructions of the at least one set of instructions.

[0022] Thirdly, this specification provides a pet vital signs detection system, which includes the aforementioned pet vital signs detection equipment and a server, with the server communicating with the detection equipment.

[0023] In some embodiments, the detection system further includes a terminal device. The terminal device is communicatively connected to the server.

[0024] As can be seen from the above technical solutions, the pet vital signs detection method, detection equipment, and detection system provided in this manual can perform routine health checks on pets, detect abnormalities in the early stages of a pet's illness in a timely manner, and help reduce the damage suffered by the pet. Determining the target pet's vital signs after confirming its location reduces computational load and improves the accuracy and reliability of the detection results. Inertial sensors can directly or indirectly contact the target pet, enabling contact-based detection. Contact-based detection requires less workspace and is suitable for detecting target pets in smaller spaces. Through the detection data from the inertial sensors, the detection component can determine both the target pet's location and its vital signs, thereby reducing the types of sensors required and simplifying the detection method.

[0025] Other functions of the methods, equipment, and systems for detecting pet vital signs provided in this manual will be partially listed in the following description. The figures and examples described below will be readily apparent to those skilled in the art. The inventive aspects of the methods, equipment, and systems for detecting pet vital signs provided in this manual can be fully understood through practice or use of the methods, apparatus, and combinations described in the detailed examples below. Attached Figure Description

[0026] To more clearly illustrate the technical solutions in the embodiments of this specification, the accompanying drawings used in the description of the embodiments will be briefly introduced below. Obviously, the accompanying drawings described below are only some embodiments of this specification. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0027] Figure 1 The diagram illustrates application scenarios of a pet vital signs detection system provided according to some embodiments of this specification;

[0028] Figure 2 A schematic block diagram of the circuit structure of a pet vital signs detection device provided according to some embodiments of this specification is shown; and

[0029] Figure 3 A flowchart is shown of a method for detecting vital signs of pets according to some embodiments of this specification. Detailed Implementation

[0030] The following description provides specific application scenarios and requirements for this specification, intended to enable those skilled in the art to make and use the contents of this specification. Various partial modifications to the disclosed embodiments will be apparent to those skilled in the art, and the general principles defined herein can be applied to other embodiments and applications without departing from the spirit and scope of this specification. Therefore, this specification is not limited to the embodiments shown, but rather to the widest scope consistent with the claims.

[0031] The terminology used herein is for the purpose of describing particular exemplary embodiments only and is not restrictive. For example, unless the context clearly indicates otherwise, the singular forms “a,” “an,” and “the” as used herein may also include the plural forms. When used in this specification, the terms “comprising,” “including,” and / or “containing” mean that the associated feature, integer, step, operation, element, and / or component is present, but do not preclude the presence of one or more other features, integers, steps, operations, elements, components, and / or groups, or that other features, integers, steps, operations, elements, components, and / or groups may be added to the system / method.

[0032] Considering the following description, these and other features of this specification, as well as the operation and function of the related components of the structure, and the economy of assembly and manufacture of the parts, can be significantly improved. All of these form part of this specification with reference to the accompanying drawings. However, it should be clearly understood that the drawings are for illustrative and descriptive purposes only and are not intended to limit the scope of this specification. It should also be understood that the drawings are not drawn to scale.

[0033] The flowcharts used in this specification illustrate operations implemented according to some embodiments of this specification. It should be clearly understood that the operations in the flowcharts may not be implemented in a sequential order. Instead, the operations may be implemented in reverse order or simultaneously. Furthermore, one or more additional operations may be added to the flowcharts. One or more operations may be removed from the flowcharts.

[0034] In this specification, "X includes at least one of A, B, or C" means that X includes at least A, or X includes at least B, or X includes at least C. That is, X can include only one of A, B, and C, or any combination of A, B, and C, as well as other possible content / elements. The arbitrary combination of A, B, and C can be A, B, C, AB, AC, BC, or ABC.

[0035] In this specification, unless explicitly stated otherwise, the relationships between structures can be direct or indirect. For example, when describing "A is connected to B," unless it is explicitly stated that A and B are directly connected, it should be understood that A can be directly connected to B or indirectly connected to B. Similarly, when describing "A is on top of B," unless it is explicitly stated that A is directly above B (AB is adjacent and A is above B), it should be understood that A can be directly above B or indirectly above B (AB is separated by other elements, and A is above B). And so on.

[0036] With economic and social development and the continuous improvement of people's living standards, the popularity of pets, represented by cats and dogs, in households is rapidly increasing. Consequently, people are paying more and more attention to their pets' health. Because pets cannot speak, there are communication barriers between them and humans. Therefore, people often only discover abnormalities in their pets' health when their condition has become quite serious. Therefore, routine health checks for pets have become particularly important.

[0037] In view of this, embodiments of this specification provide a method, device, and system for detecting pet vital signs. Based on detection data from an inertial sensor, it is determined whether the target pet is at a preset position. If the target pet is at the preset position, its vital signs are determined based on the detection data. For example, parameters of vital signs such as breathing and heartbeat can be determined using data from the inertial sensor. This allows for routine health checks on pets, enabling the timely detection of abnormalities in the early stages of illness, thus reducing potential harm to the pet. Furthermore, the detection method, device, and system can also be used to monitor the health of pets during their recovery period after treatment.

[0038] The technical solutions of the embodiments of this specification will now be described in detail with reference to the accompanying drawings.

[0039] Figure 1 The diagram illustrates an application scenario of a pet vital signs detection system provided according to some embodiments of this specification. Figure 2 A schematic block diagram of the circuit structure of a pet vital signs detection device provided according to some embodiments of this specification is shown. Figure 1 and Figure 2As shown, the detection device 100 can perform contact-based detection of the vital signs of the target pet 400. When the target pet 400 is in a preset position 111, the detection device 100 makes direct or indirect contact with the target pet 400, thereby obtaining the parameters of the target pet 400's vital signs. The target pet 400 can be a cat, dog, or other animal. Of course, in addition to pets, the detection device 100 can also detect the vital signs of other artificially bred, raised, observed, or treated animals.

[0040] In some embodiments, the vital signs parameters of the target pet 400 may include respiratory parameters and heart rate parameters. Respiratory parameters may include respiratory rate, respiratory rhythm, and respiratory pattern. Heart rate parameters may include heart rate, heart rate rhythm, and heart rate variability.

[0041] In some embodiments, when the heart rate or respiration of the target pet 400 is higher or lower than 20% of its normal resting heart rate or respiratory rate for more than 30 minutes, the detection device 100 may issue an alarm or provide a notification in the target pet 400's vital signs report. For example, the normal resting respiratory rate range for an adult cat is 20-40 breaths / minute, and its heart rate range is 120-160 beats / minute.

[0042] In some embodiments, the detection device 100 can statistically analyze the average and extreme values ​​of the heart rate variability of a pet over a period of time (e.g., three weeks) as a benchmark, and simultaneously analyze the average and extreme values ​​of the heart rate variability on a given day. If the average and extreme values ​​of the heart rate variability on a given day deviate significantly from the benchmark average (e.g., greater than 30%), the detection device 100 can issue an alarm or provide a notification in the vital signs report of the target pet 400.

[0043] In some embodiments, the detection device 100 may perform statistics, provide prompts, or issue warnings when it detects that the target pet 400 is having difficulty breathing or has an abnormal breathing pattern.

[0044] In some embodiments, the detection device 100 can count the resting time and percentage of pets in their dens. When there is a significant abnormality (such as greater than or less than 30% of the daily average), the detection device 100 can provide corresponding prompts and warnings.

[0045] In some embodiments, the detection device 100 can continuously monitor changes in the pet's heart rate and respiratory rate, and determine changes in the pet's health status based on indicators such as stable, continuously increasing, or continuously decreasing.

[0046] like Figure 1 As shown, the detection device 100 includes a device body 110, an inertial sensor 120, and a detection component 130. The device body 110 includes a preset position 111 for accommodating the target pet 400.

[0047] In some embodiments, the main body 110 of the device may be the main part of a pet care box, pet cage, pet toilet, pet bed, or pet sleeping mat.

[0048] An inertial sensor 120 is disposed on the main body 110 of the device. When the detection device 100 is in direct or indirect contact with the target pet 400, the vibrations generated by the movement, breathing, and heartbeat of the target pet 400 can be transmitted to the inertial sensor 120, thereby enabling the inertial sensor 120 to obtain detection data reflecting the vital signs and movement of the target pet 400. The inertial sensor 120 can continuously obtain detection data. During the process of the inertial sensor 120 obtaining detection data, the target pet 400 may be in or away from the preset position 111, so the detection data of the inertial sensor 120 can also reflect the position of the target pet 400.

[0049] The inertial sensor 120 can be disposed at the bottom of the device body 110, and the vibrations generated by the target pet 400 can be transmitted to the inertial sensor 120 through the bottom of the device body 110. For example, the bottom of the detection device 100 can be provided with a wooden board, a sponge pad, or a spring pad, and the inertial sensor 120 can be located below the wooden board, sponge pad, or spring pad, or it can be suspended below the inertial sensor 120. In this way, the bottom of the detection device 100 can fully transmit the vibration to the inertial sensor 120, and the inertial sensor 120 itself can also vibrate with the movement of the target pet 400.

[0050] The inertial sensor 120 may include a gyroscope and an accelerometer. The gyroscope is capable of measuring the angular velocity of an object in multiple degrees of freedom. The accelerometer is capable of measuring the acceleration of an object in multiple degrees of freedom. Therefore, the inertial sensor 120 is capable of measuring the motion of the object being measured in multiple degrees of freedom.

[0051] The detection data obtained by the inertial sensor 120 can be arranged according to the temporal and data dimensions, and can be presented in matrix form. Specifically, the inertial sensor 120 can acquire one frame of data at a given time point, and the data from different frames can be arranged in temporal order. Each frame of data acquired by the inertial sensor 120 can also be arranged according to the data dimension. In the data dimension, the data from different frames can correspond to each other. For example, the detection data can be represented as follows, where i = 0, 1, ..., n-1, n is the number of data dimensions, j = 0, 1, ..., t-1, and t is the number of frames of the detection data. The number of frames of detection data obtained by the inertial sensor 120 can be 10 frames / second or 20 frames / second.

[0052] In some embodiments, each data dimension may correspond to one degree of freedom of motion.

[0053] like Figure 1 and Figure 2As shown, the detection component 130 is communicatively connected to the inertial sensor 120. The detection component 130 includes at least one processor 132 and at least one storage medium 131.

[0054] At least one storage medium 131 stores at least one set of instructions for detecting the vital signs of the target pet 400. Storage medium 131 may include one or more of a disk, a read-only storage medium, or a random access storage medium. Storage medium 131 may also include non-volatile random access memory.

[0055] At least one processor 132 is communicatively connected to at least one storage medium 131. When the detection device 100 is running, the at least one processor 132 reads at least one instruction set and executes the following detection method according to the instructions of the at least one instruction set.

[0056] Processor 132 may be in the form of one or more processors 132. According to some embodiments of this specification, processor 132 may include one or more hardware processors, such as microcontrollers, microprocessors, reduced instruction set computers (RISC), application-specific integrated circuits (ASICs), application-specific instruction set processors (ASIPs), central processing units (CPUs), microprocessors (MCUs), graphics processing units (GPUs), physical processing units (PPUs), microcontroller units, digital signal processors (DSPs), field-programmable gate arrays (FPGAs), advanced RISC machines (ARMs), programmable logic devices (PLDs), any circuit or processor capable of performing one or more functions, etc., or any combination thereof.

[0057] For illustrative purposes only, only one processor 132 is described in this specification for the detection device 100. However, it should be noted that the detection device 100 may also include multiple processors 132, and this specification does not limit the number of processors 132 included in the detection device 100. Therefore, the operation and / or method steps disclosed in this specification can be executed by one processor 132 as described in this specification, or they can be executed jointly by multiple processors 132. For example, if the processor 132 of the detection device 100 in this specification executes steps A and B, it should be understood that steps A and B can also be executed jointly or separately by two different processors 132 (e.g., step A is executed by the first processor 132, step B is executed by the second processor 132, or steps A and B are executed jointly by the first and second processors 132).

[0058] In some embodiments, the processor 132 is also used for overall control of the detection device 100, acquisition and processing of detection data obtained by the inertial sensor 120, and transmission of the processing results of the detection data to the outside via the communication component 140. The storage medium 131 can also be used for storing the detection data and the processing results of the detection data. In some embodiments, the processing results of the detection data include health indicator data of the target pet 400 and the detection results of the vital signs of the target pet 400.

[0059] In some embodiments, the detection device 100 may notify the user after obtaining the health indicator data and / or vital sign detection results of the target pet 400. In some embodiments, the detection device 100 may notify the user if the vital sign detection results of the target pet 400 indicate an abnormality in the target pet 400. Further, the detection device 100 may use the language module, electronic display screen, or LED lights of the device body 110 to notify the user of the aforementioned health indicator data and / or vital sign detection results of the target pet 400 in the form of numbers, text, charts, reports, or lights.

[0060] In some embodiments, the detection device 100 may further include an internal communication bus disposed on the device body 110. The internal communication bus can connect different system components, including storage medium 131, processor 132, and inertial sensor 120. Meanwhile, the device body 110 may also include I / O components. The I / O components support input / output between the device body 110 and other components.

[0061] In some embodiments, the detection device 100 may further include a communication component 140 disposed on the device body 110, the communication component 140 being used for data communication between the device body 110 and the outside world. For example, the communication component 140 may be used for data communication between the device body 110 and a network. The communication component 140 may be a wired communication component 140 or a wireless communication component 140. For example, the communication component 140 may include a Bluetooth module, a Wi-Fi module, etc.

[0062] like Figure 1 and Figure 2 As shown, the pet vital signs detection system 1 includes a detection device 100 and a server 200, with the server 200 communicatively connected to the detection device 100. Furthermore, the detection system 1 also includes a terminal device 300, which is communicatively connected to the server 200.

[0063] In some embodiments, the detection component 130 is disposed on the device body 110. The detection device 100 can independently detect the vital signs of the target pet 400 without relying on other devices. In other embodiments, the detection component 130 is at least partially disposed on the terminal device 300 or the server 200. The detection device 100 operates as an accessory or functional component of the detection system 1, such as a pet care box, pet cage, pet toilet, pet feeder, etc. Detection data obtained by the inertial sensor 120 can be transmitted to the terminal device 300 or the server 200 via the communication component 140.

[0064] In some embodiments, the terminal device 300 and the detection device 100 can be wirelessly connected via Bluetooth, or wired connected via a physical medium (such as a telephone line, Ethernet cable, coaxial cable, optical fiber, etc.).

[0065] In some embodiments, the detection device 100 can send detection data to the terminal device 300, which then analyzes or performs other further processing on the data. In some embodiments, the detection device 100 can send the vital sign detection results of the target pet 400 to the terminal device 300, which can then directly display these results for user viewing. In some embodiments, the detection device 100 can send both the health indicator data and the vital sign detection results of the target pet 400 to the terminal device 300 for user viewing. In some embodiments, the detection device 100 can send a reminder message to the terminal device 300 if the vital sign detection results indicate an abnormality in the target pet 400's vital signs, thus alerting the user. In some embodiments, the terminal device 300 can control the activation and deactivation of the detection device 100. This approach saves power consumption of the detection device 100, facilitates user operation, and provides a superior user experience.

[0066] In this specification, terminal device 300 may include mobile devices, tablets, laptops, built-in devices in motor vehicles, or similar content, or any combination thereof. In some embodiments, the mobile device may include smart home devices, smart mobile devices, virtual reality devices, augmented reality devices, or similar devices, or any combination thereof. In some embodiments, the smart home device may include smart TVs, desktop computers, etc., or any combination thereof. In some embodiments, the smart mobile device may include smartphones, personal digital assistants, gaming devices, navigation devices, etc., or any combination thereof. In some embodiments, the virtual reality device or augmented reality device may include virtual reality headsets, virtual reality glasses, virtual reality patches, augmented reality headsets, augmented reality glasses, augmented reality patches, or similar content, or any combination thereof. For example, the virtual reality device or the augmented reality device may include AR glasses, head-mounted displays, VR, etc. In some embodiments, the built-in device in the motor vehicle may include an in-vehicle computer, an in-vehicle television, etc.

[0067] In some embodiments, the terminal device 300 may have one or more applications (APPs) installed. The APP provides users with the ability and interface to interact with the outside world via a network. Specifically, the terminal device 300 may have a target APP installed, which provides users with a display interface and control interface related to the pet vital sign detection method described in this specification.

[0068] In some embodiments, the terminal device 300 can provide users with a display interface and control interface related to the pet vital signs detection method described in this specification through a mini-program.

[0069] In some embodiments, the detection device 100 can send the health indicator data of the target pet 400 to the server 200, which then analyzes and / or performs other further processing on the health indicator data of the target pet 400 to obtain the detection results of vital signs. In some embodiments, the detection device 100 can send the health indicator data of the target pet 400 and the detection results of vital signs to the server 200.

[0070] Furthermore, the server 200 can send the obtained vital sign detection results of the target pet 400 to the terminal device 300. Through the detection device 100, the server 200, and the terminal device 300, a remote monitoring system for pet vital signs can be formed, allowing users to remotely view the vital sign detection results of the target pet 400 even when they are not near it, via the terminal device 300.

[0071] This specification, in another aspect, provides a non-transitory storage medium storing at least one set of executable instructions for detecting pet vital signs. When the executable instructions are executed by a processor, they instruct the processor to perform the steps of the pet vital sign detection method described in this specification. In some possible embodiments, various aspects of this specification can also be implemented as a program product comprising program code. When the program product is run on a detection device, the program code causes the detection device to perform the steps of the pet vital sign detection method described in this specification. The program product for implementing the above method may employ a portable compact disc read-only memory (CD-ROM) containing program code and may run on a detection device. However, the program product of this specification is not limited thereto. In this specification, the readable storage medium may be any tangible medium containing or storing a program that can be used by or in conjunction with an instruction execution system. The program product may employ any combination of one or more readable media. The readable medium can be a readable signal medium or a readable storage medium, such as: a portable disk, a hard disk, random access memory (RAM), read-only memory (ROM), erasable programmable read-only memory (EPROM or flash memory), portable compact disk read-only memory (CD-ROM), optical storage device, magnetic storage device, or any suitable combination thereof. The program code contained on the readable storage medium can be transmitted using any suitable medium, including but not limited to wireless, wired, optical fiber, RF, etc., or any suitable combination thereof. The program code for performing the operations of this specification can be written in any combination of one or more programming languages, including object-oriented programming languages—such as Java, C++, etc.—and conventional procedural programming languages—such as the "C" language or similar programming languages. The program code can be executed entirely on the testing device, partially on the testing device, as a standalone software package, partially on the testing device and partially on a remote computing device, or entirely on a remote computing device.

[0072] The above are illustrative schemes of a pet vital signs detection device and a detection system according to embodiments of this specification. It should be noted that the illustrative schemes of the pet vital signs detection device and the detection system are based on the same concept as the detection method described below. Details in the illustrative schemes of the pet vital signs detection device, the detection system, and the detection method can be referred to interchangeably.

[0073] Figure 3A flowchart of a pet vital signs detection method P100 according to some embodiments of this specification is shown. The detection method can be executed by a detection component 130. Specifically, a processor 132 is communicatively connected to the inertial sensor 120, the communication component 140, and the storage medium 131. By reading the instruction set in the storage medium 131 and combining it with the data detected by the inertial sensor 120, the processor executes the steps in flowchart P100. For ease of description, the entity executing the detection method will be referred to as the detection component.

[0074] like Figure 3 As shown, the method for detecting vital signs in pets, P100, includes the following steps:

[0075] S100: Determines whether the target pet is in a preset position based on detection data from the inertial sensor.

[0076] S200: If the target pet is in a preset location, determine the target pet's vital signs based on the detection data.

[0077] The target pet's activity area may include a preset location. In some embodiments, the preset location refers to the interior of a pet care box, pet cage, pet toilet, pet bed, or the area above a pet sleeping mat.

[0078] When the target pet is in the preset position, the vibrations generated by the target pet can be transmitted to the inertial sensor 120 more effectively, and the detection data of the inertial sensor 120 will more accurately reflect the vital signs of the target pet. When the target pet leaves the preset position, the detection data of the inertial sensor 120 is not very meaningful in reflecting the vital signs of the target pet. Therefore, by determining that the target pet is in the preset position before determining the vital signs of the target pet, the detection component 130 can reduce the amount of computation and improve the accuracy and reliability of the detection results of vital signs.

[0079] The inertial sensor 120 can directly or indirectly contact the target pet, thus enabling contact detection. Indirect contact refers to contact through a solid material that can transmit vibrations. For non-contact detection, the sensor typically needs to be at a certain distance from the target pet. Compared to non-contact detection, contact detection requires less workspace and is suitable for detecting target pets in smaller spaces.

[0080] By using the detection data from the inertial sensor 120, the detection component 130 can determine the location of the target pet and its vital signs, thereby reducing the number of sensor types required and simplifying the detection method.

[0081] In some embodiments, in order to determine whether the target pet is in a preset position based on the detection data from the inertial sensor 120, step S100 can be divided into the following sub-steps S110, S120, S130 and S140.

[0082] S110: Receives detection data from the inertial sensor.

[0083] In some embodiments, the detection data from the inertial sensor 120 can be directly transmitted to the detection component 130. In some embodiments, the detection data can be transmitted to the detection component 130 via a server or terminal device. In other embodiments, the detection data can be stored in a storage medium located outside the detection component 130 and retrieved by the detection component 130.

[0084] S120: Obtain the current signal sequence based on the detection data, wherein the current signal sequence includes the signal of the current frame and the signals of multiple historical frames that are in time before the current frame.

[0085] The signal of the current frame is the signal obtained by the detection component 130 at the present moment, and the signal of the historical frame is the signal obtained by the detection component 130 at a past moment. The current frame and the historical frames can be arranged in chronological order. The historical frame can be regarded as the previous current frame, and the current frame can be regarded as the future historical frame.

[0086] In some embodiments, the detection data includes initial current data and multiple frames of initial historical data that precede the initial current data in time sequence. Both the initial current data and each frame of initial historical data have multiple data dimensions.

[0087] For example, the detection data can be represented as s i,j Where i = 0, 1, ..., n-1, n is the number of data dimensions, j = 0, 1, ..., t-1, t is the number of frames of the detected data. The initial current data can be represented as s. i , where i = 0, 1, ..., n-1, and n is the number of data dimensions.

[0088] The specific implementation of obtaining the current signal sequence based on the detection data may further include sub-steps S121 and S122.

[0089] S121: In each data dimension, calculate the average value of the initial historical data of the first preset number of frames adjacent to the initial current data.

[0090] The initial historical data for the first preset frame number is the initial historical data obtained by the inertial sensor 120 within a certain time window before the current moment. For example, the first preset frame number is denoted as N, where N = 100, 200, or 300.

[0091] The average value of the initial historical data for the first preset frame number N is

[0092] S122: Subtract the average value from the initial current data in the corresponding data dimension to obtain the signal of the current frame, and then use the signal of the current frame and the signals of multiple historical frames to form the current signal sequence.

[0093] That is, s i g i =s i -bg i , where s i g i The signal for the current frame is i = 0, 1, ..., n-1.

[0094] The initial current data refers to the data detected by the inertial sensor 120 at the current moment, while the initial historical data refers to the data detected by the inertial sensor 120 at past moments. In terms of timing, the initial current data corresponds to the signal of the current frame. During operation, the inertial sensor 120 is subject to noise interference from background signals, which are reflected in the detected data. Therefore, the detection component 130 needs to remove the background from the initial current data to obtain the signal of the current frame and the historical frame signals. This improves the reliability of the feature parameters extracted in subsequent processes.

[0095] The average value calculated by the detection component 130 in step S121 can reflect the background signal. That is... It can be used as the background signal of the current frame signal. Through step S122, the background signal can be subtracted from the initial current data to reduce the adverse effects caused by noise in the background signal.

[0096] S130: Perform feature extraction on the current signal sequence to obtain the feature parameters corresponding to the current frame.

[0097] In some embodiments, the feature parameters corresponding to the current frame may include a feature vector. The signal of the current frame and the signal of each historical frame have multiple data dimensions. Feature extraction is performed on the current signal sequence to obtain the feature parameters corresponding to the current frame, including:

[0098] S131: Extract features from the current signal sequence to obtain multiple signal features related to the current frame, and combine the multiple signal features into a feature vector.

[0099] For example, the feature vector V = (v1, v2, ..., vk), where k is the number of multiple signal features.

[0100] The detection component 130 utilizes multi-dimensional feature vectors to train location detection and activity detection models, as well as to detect the location and movement of pets. This simplifies model training and improves detection accuracy. Multiple signal features can be reduced or expanded based on the performance of the hardware and software platform during implementation.

[0101] In some embodiments, the multiple signal features include features of the current frame. Feature extraction is performed on the current signal sequence to obtain multiple signal features related to the current frame, including:

[0102] S132: Perform feature extraction on the signal of the current frame to obtain the features of the current frame. The features of the current frame include at least one of the mean, maximum, minimum, median, or standard deviation of the signal. The signal of the current frame has multiple data dimensions and can be considered as a one-dimensional vector; therefore, the signal s of the current frame can be calculated. i g i The mean, maximum, minimum, median, or standard deviation of the included data.

[0103] In some embodiments, feature extraction is performed on the current signal sequence to obtain multiple signal features related to the current frame, including:

[0104] S133: Calculate the absolute value of the difference between the signal of the current frame and the historical frame signal of the adjacent current frame in each data dimension to obtain the differential signal.

[0105] That is, the differential signal diff i =|s i g i,j -s i g i,j-1 |, where i = 0, 1, ..., n-1, and j is the current frame.

[0106] S134: Perform feature extraction on the differential signal to obtain the features of the differential signal. These features include multiple signal features, specifically the features of the differential signal, which include at least one of the mean, maximum, minimum, median, or standard deviation of the differential signal.

[0107] Differential signals can reflect the pet's movement and position changes at any given moment. Corresponding to the signals of the current frame and historical frames, the differential signal also has multiple data dimensions and can be viewed as a one-dimensional vector; therefore, the differential signal diff can be calculated. i The mean, maximum, minimum, median, or standard deviation of the included data.

[0108] In some embodiments, feature extraction is performed on the current signal sequence to obtain multiple signal features related to the current frame, including:

[0109] S135: Combine the signal of the current frame with the historical frame signal of the second preset frame number adjacent to the current frame to form the first sub-signal sequence.

[0110] The historical frame signal of the second preset frame number is the historical frame signal obtained by the detection component 130 within a certain time window before the current moment. For example, the second preset frame number is denoted as m, where m = 100, 200, or 300. The first sub-signal sequence is denoted as s. i g i,j , where i = 0, 1, ..., n-1, j = 0, 1, ..., m-1.

[0111] S136: Perform feature extraction on the first sub-signal sequence to obtain the features of the first sub-signal sequence. The features include multiple signal features, including the features of the first sub-signal sequence, and the features of the first sub-signal sequence include at least one of the standard deviation or covariance of the first sub-signal sequence in each data dimension.

[0112] The first sub-signal sequence s i g i,j This is a subset of the current signal sequence. The characteristics of the first sub-signal sequence reflect the changes in the pet's movement and position within a certain time window before the current moment. Corresponding to the signals of the current frame and historical frames, the differential signal has multiple data dimensions. Therefore, the first sub-signal sequence s can be calculated. ii g i,j The standard deviation or covariance of the data contained in each data dimension.

[0113] In some embodiments, the multiple signal features include confidence features. Specific implementations of feature extraction on the current signal sequence to obtain multiple signal features related to the current frame can be found below, including steps S137, S138, and S139.

[0114] S137: Combine the signal of the current frame with the historical frame signal of the third preset frame number adjacent to the current frame to form a second sub-signal sequence.

[0115] The third preset frame number is the historical frame signal obtained by the detection component 130 within a certain time window before the current moment. For example, the third preset frame number is denoted as p, where p = 100, 200, or 300. The second sub-signal sequence is denoted as s. i,j , where i = 0, 1, ..., n-1, j = 0, 1, ..., p-1.

[0116] S138: Calculate the credibility of each data dimension based on the second sub-signal sequence.

[0117] For the second sub-signal sequence, the greater the credibility value of each data dimension, the greater the credibility value of the second sub-signal sequence, that is, the higher the reliability and accuracy of the data of the second sub-signal sequence in reflecting the vital signs of the pet. Since the detection component 130 continuously and long-term detects the vital signs of the pet, the detection component 130 can obtain multiple second sub-signal sequences. The detection component 130 can use the data of the time window corresponding to the second sub-signal sequence with a higher credibility (higher than the credibility threshold) to determine the vital sign parameters of the pet, and discard the data of the time window corresponding to the second sub-signal sequence with a smaller credibility (lower than the credibility threshold).

[0118] For each data dimension, the greater the credibility value of the data dimension, the higher the reliability and accuracy of the data of the data dimension in reflecting the vital signs of the pet. The detection component 130 can use the data of the data dimension with a higher credibility (such as higher than the credibility threshold) to determine the movement situation and position situation of the pet, and discard the data of the data dimension with a smaller credibility (such as lower than the credibility threshold). This can improve the accuracy of detecting vital sign parameters, simplify the feature extraction process, and reduce the amount of computation.

[0119] In some embodiments, the vital signs include respiratory parameters. The specific implementation manner of calculating the credibility of each data dimension based on the second sub-signal sequence can refer to the following content including steps S1381, S1382, S1383, S1384, S1385, and S1386.

[0120] S1381: In each data dimension, perform a smoothing operation on the data of the second sub-signal sequence to obtain a smoothed data sequence corresponding to the data dimension.

[0121] For example, there is a data dimension x, 0 < x < n, and the data of the x-th data dimension in the second sub-signal sequence is denoted as s x,j . The detection component 130 can perform a smoothing operation on s x,j to obtain a smoothed data sequence S ′ x,j .

[0122] In some embodiments, the detection component 130 can perform a smoothing operation on s x,j by means of mean filtering or Gaussian filtering.

[0123] S1382: Perform a fast Fourier transform on the smoothed data sequence to obtain an actual frequency domain signal.

[0124] For example, the detection component 130 can perform a fast Fourier transform (FFT) with a length of M (such as M = 1024) on the smoothed data sequence S ′ x,j to obtain an actual frequency domain signal freqm Where m = 0, 1, ..., M / 2.

[0125] The signal obtained by the detection component 130 through the inertial sensor 120 is a time-domain signal. Vital signs such as breathing and heartbeat have periodicity and regularity. The detection component 130 can convert the time-domain signal into a frequency-domain signal through step S1382, which can reveal the spectral characteristics of the detected signal and realize rapid analysis and processing of the signal.

[0126] S1383: Perform full-domain peak finding on the actual frequency domain signal to obtain the actual maximum peak value. The actual maximum peak value can be denoted as pk0.

[0127] S1384: Find the target maximum peak value and the target second largest peak value within a preset interval for the actual frequency domain signal. The preset interval corresponds to the respiratory parameters.

[0128] For example, the preset interval is the range of the frequency domain signal corresponding to the theoretical respiratory rate of the target pet, which is m1 to m2. The target's maximum peak value can be denoted as pk1, and the target's second largest peak value can be denoted as pk2. Among them, pk1 > pk2.

[0129] S1385: Compare the actual maximum peak value with the target maximum peak value.

[0130] S1386: If the actual maximum peak value equals the target maximum peak value, then the credibility of the corresponding data dimension is equal to the ratio of the target maximum peak value to the target second-largest peak value. If the actual maximum peak value does not equal the target maximum peak value, then the credibility of the corresponding data dimension is equal to the ratio of the actual maximum peak value to the target maximum peak value.

[0131] The credibility of the i-th data dimension can be denoted as conf. i When pk0 = pk1, conf i =pk1 / pk2; otherwise, conf i =pk0 / pk1.

[0132] If the actual maximum peak value pk0 equals the target maximum peak value pk1, converting the frequency corresponding to the actual maximum peak value pk0 yields the respiratory rate value. In this case, the correlation between the actual frequency domain signal and respiratory parameters is strong, and the reliability of the corresponding data dimension is high. i The value is relatively large. For example, conf i >1.

[0133] If the actual maximum peak value pk0 is not equal to the target maximum peak value pk1, it is difficult to obtain the respiratory rate value from the actual maximum peak value pk0. In this case, the correlation between the actual frequency domain signal and respiratory parameters is weak, and the reliability of the corresponding data dimension is low. i The value is small. For example, confi <1.

[0134] Therefore, through steps S1381, S1382, S1383, S1384, S1385, and S1386, the detection component can obtain the reliability of the data dimensions for respiratory parameters. That is, conf i The larger the value, the stronger the correlation between the corresponding data dimension and the respiratory parameters, and the higher the reliability and accuracy of the determined respiratory parameters.

[0135] When determining the reliability of data dimensions related to respiratory parameters, bandpass filtering of the respiratory frequency band is not performed, thus avoiding interference from bandpass filtering on the determined reliability. Bandpass filtering of the respiratory frequency band during the determination of respiratory parameters can amplify respiratory-related features, potentially distorting the determined respiratory parameters. The higher the reliability of the data dimension for respiratory parameters, the less likely the determined respiratory parameters are to be distorted. Therefore, through conf... i Respiratory parameters determined by data dimensions with larger values ​​are more accurate and reliable. This can improve the accuracy of vital sign parameter detection, simplify the feature extraction process, and reduce computational load.

[0136] S139: Use at least one of the maximum, mean, or median values ​​of the confidence level for each data dimension as a confidence level feature.

[0137] Through steps S1381, S1382, S1383, S1384, S1385, and S1386, the detection component 130 can obtain the confidence level (conf) for each of multiple data dimensions. i Thus, component 130 can calculate the credibility of each data dimension. i The maximum, mean, or median value.

[0138] S140: Input the feature parameters into the pre-trained location detection model to determine whether the target pet is in the preset location.

[0139] Through step S140, the detection component 130 can determine whether the target pet is in a preset position or has left the preset position.

[0140] In some embodiments, before inputting the feature parameters into a pre-trained location detection model, the following is included:

[0141] S101: Receive training data from the inertial sensor. The training data is measured by the inertial sensor when the target pet is in a preset position and when it leaves the preset position.

[0142] S102: Extract features from the training data to obtain training feature parameters of the same type as the feature parameters.

[0143] During the training phase of the position detection model, the data acquired by the inertial sensor 120 serves as the training data. The training data and the detection data are data acquired by the inertial sensor 120 at different stages. The detection component 130 obtains the training feature parameters in the same way as the feature parameters. For example, the training feature parameters may include at least one of the following: features of the current frame, features of the differential signal, features of the first sub-signal sequence, or confidence features.

[0144] S103: Train the location detection model using training feature parameters.

[0145] In some embodiments, the detection component 130 may train the location detection model using common machine learning methods. For example, the detection component 130 may train the location detection model using decision trees or random forests. In other embodiments, the detection component 130 may train the location detection model using deep learning.

[0146] In some embodiments, before determining the vital signs of the target pet based on the detection data, the method includes: S150: determining whether the target pet is in a resting state based on the detection data. If the target pet is in a preset position and in a resting state, then the vital signs are determined based on the detection data.

[0147] The target pet can be in an active or resting state. When the target pet is active, the vibrations generated by its movements are recorded by the inertial sensor 120, resulting in low accuracy of the vital sign parameters obtained by the detection component 130. Therefore, this embodiment determines the vital sign parameters of the target pet when it is in a resting state, thereby improving the accuracy of the target pet's vital sign detection.

[0148] In some embodiments, determining whether a target pet is in a resting state based on detection data includes:

[0149] S151: Input the feature parameters into the pre-trained activity detection model to determine whether the target pet is in a resting state.

[0150] Through step S151, the detection component 130 can determine whether the target pet is in a resting or active state. Furthermore, the detection component 130 can determine the activity level of the active state based on the detection data.

[0151] In some embodiments, before inputting the feature parameters into a pre-trained activity detection model, the following is included:

[0152] S104: Train the activity detection model using training feature parameters.

[0153] In some embodiments, the detection component 130 may train the activity detection model using common machine learning methods. For example, the detection component 130 may train the activity detection model using decision trees or random forests. In other embodiments, the detection component 130 may train the activity detection model using deep learning.

[0154] In some embodiments, vital signs include respiratory parameters. The current signal sequence has multiple data dimensions. Determining the vital signs of the target pet based on the detection data includes:

[0155] S210: Within the first preset time window, obtain the signal of the data dimension with the highest confidence in the current signal sequence to form a respiratory signal sequence.

[0156] For example, the first preset time window can be 20 seconds, 30 seconds, or 40 seconds. Using confidence level as the basis for selecting signal dimensions can simplify the detection process and improve the accuracy of extracting heartbeat and respiratory signs.

[0157] S220: Perform bandpass filtering on the respiratory signal sequence in a first preset frequency band to obtain the respiratory waveform signal of the target pet. The first preset frequency band is the pet's respiratory frequency band.

[0158] S230: Determining respiratory parameters based on respiratory waveform signals. These respiratory parameters include at least one of respiratory rate, respiratory rhythm, or respiratory pattern.

[0159] Through steps S210, S210, and S210, the detection component 130 can determine the respiratory parameters. For example, the detection component 130 can determine the respiratory rate. The respiratory rate is the frequency of breathing. The respiratory rate of the target pet may differ between healthy and abnormal conditions. Therefore, by monitoring the respiratory rate of the target pet, the target pet's physical condition can be monitored. When the target pet breathes, its body vibrates rhythmically with the breathing pattern, and this vibration is transmitted to the inertial sensor 120. Therefore, by detecting the respiratory waveform signal, the respiratory rate of the target pet can be determined.

[0160] In some embodiments, vital signs include heart rate parameters. Determining the vital signs of a target pet based on detection data further includes:

[0161] S240: Differentiate the respiratory signal sequence with the respiratory waveform signal to obtain the heartbeat signal sequence.

[0162] S250: Perform bandpass filtering on the heartbeat signal sequence in the second preset frequency band to obtain the heartbeat waveform signal of the target pet. The second preset frequency band is the heartbeat frequency band of the pet.

[0163] S260: Determining heartbeat parameters based on the heartbeat waveform signal. The heartbeat parameters include at least one of heart rate or heart rate variability.

[0164] Heart rate is the frequency of heartbeats. Heart rate variability (HRV) refers to the variation in the difference between successive heartbeat cycles. The heart rate parameters of a target pet may differ depending on whether it is in a healthy or abnormal condition. Therefore, monitoring the target pet's heart rate parameters can monitor its physical condition. When the target pet's heart beats, its body vibrates rhythmically with the heartbeat, and this vibration is transmitted to the inertial sensor 120. Therefore, the detection component 130 can detect the target pet's heart rate parameters through the heartbeat waveform signal.

[0165] Since the vibration caused by the target pet's breathing is much greater than the vibration caused by its heartbeat, it is necessary to eliminate the influence of the vibration caused by the target pet's breathing on the detected heartbeat parameters through step S240.

[0166] In other embodiments, the detection component 130 may compare the detection data with a parameter threshold corresponding to the pet's location to determine whether the target pet is in a preset location.

[0167] Furthermore, the detection component 130 can compare the detection data with the parameter thresholds corresponding to the pet's movement to determine whether the target pet is in a resting state.

[0168] In summary, the detection method provided in this application can determine the target pet's vital signs only after confirming that the target pet is in a preset position, thereby reducing computational load and improving the accuracy and reliability of the vital sign detection results. The inertial sensor can directly or indirectly contact the target pet, thus enabling contact-based detection. Contact-based detection requires less workspace, making it suitable for detecting target pets in smaller spaces. Through the detection data from the inertial sensor, the detection component 130 can determine both the target pet's position and its vital signs, thereby reducing the number of sensor types required and simplifying the detection method.

[0169] The foregoing has described specific embodiments of this specification. Other embodiments are within the scope of the appended claims. In some cases, the actions or steps recited in the claims may be performed in a different order than that shown in the embodiments and may still achieve the desired result. Furthermore, the processes depicted in the drawings do not necessarily require a specific or sequential order to achieve the desired result. In some embodiments, multitasking and parallel processing are possible or may be advantageous.

[0170] In summary, after reading this detailed disclosure, those skilled in the art will understand that the foregoing detailed disclosure is presented by way of example only and is not restrictive. Although not explicitly stated herein, those skilled in the art will understand that this specification requires various reasonable changes, improvements, and modifications to the embodiments. These changes, improvements, and modifications are intended to be made by this specification and are within the spirit and scope of the exemplary embodiments described herein.

[0171] Furthermore, certain terms in this specification have been used to describe embodiments of this specification. For example, "an embodiment," "an embodiment," and / or "some embodiments" mean that a particular feature, structure, or characteristic described in connection with that embodiment may be included in at least one embodiment of this specification. Therefore, it is to be emphasized and understood that two or more references to "an embodiment" or "an embodiment" or "alternative embodiment" in various parts of this specification do not necessarily refer to the same embodiment. Moreover, specific features, structures, or characteristics may be suitably combined in one or more embodiments of this specification.

[0172] It should be understood that in the foregoing description of the embodiments in this specification, various features are combined in a single embodiment, drawing, or description for the purpose of simplifying the description and aiding in the understanding of a feature. However, this does not mean that the combination of these features is necessary, and those skilled in the art may readily identify some of the devices as separate embodiments when reading this specification. That is, the embodiments in this specification can also be understood as an integration of multiple secondary embodiments. It is also valid when each secondary embodiment contains fewer than all the features of a single foregoing disclosed embodiment.

[0173] Every patent, patent application, publication of a patent application, and other material, such as articles, books, specifications, publications, documents, and literature (excluding any related historical examination documents), cited in this disclosure is incorporated herein for all purposes, including, for example, in the specification and claims of this disclosure. However, in the event of any inconsistency or conflict between the descriptions, definitions, and / or terms used in the foregoing and those used in this disclosure, the descriptions, definitions, and / or terms used in this disclosure shall prevail.

[0174] Finally, it should be understood that the embodiments disclosed herein are illustrative of the principles of the embodiments described in this specification. Other modified embodiments are also within the scope of this specification. Therefore, the embodiments disclosed in this specification are merely examples and not limitations. Those skilled in the art can implement the applications described in this specification using alternative configurations based on the embodiments in this specification. Therefore, the embodiments in this specification are not limited to the embodiments precisely described in the applications.

Claims

1. A method for detecting vital signs in pets, characterized in that, include: Determine whether the target pet is in the preset position based on detection data from inertial sensors; as well as If the target pet is in the preset position, the vital signs of the target pet are determined based on the detection data.

2. The method as described in claim 1, characterized in that, The step of determining whether the target pet is in a preset position based on detection data from an inertial sensor includes: Receive detection data from the inertial sensor; The current signal sequence is obtained based on the detection data, wherein the current signal sequence includes the signal of the current frame and the signals of multiple historical frames that are timed before the current frame. Feature extraction is performed on the current signal sequence to obtain feature parameters corresponding to the current frame; and The feature parameters are input into a pre-trained location detection model to determine whether the target pet is in the preset location.

3. The method as described in claim 2, characterized in that, Before inputting the feature parameters into the pre-trained location detection model, the following steps are included: Receive training data from the inertial sensor; wherein the training data is measured by the inertial sensor when the target pet is in the preset position and when it leaves the preset position; Feature extraction is performed on the training data to obtain training feature parameters of the same type as the feature parameters; and The location detection model is trained using the training feature parameters.

4. The method as described in claim 2, characterized in that, Before determining the vital signs of the target pet based on the detection data, the procedure includes: determining whether the target pet is in a resting state based on the detection data; If the target pet is in the preset location and in the resting state, the vital signs are determined based on the detection data.

5. The method as described in claim 4, characterized in that, The step of determining whether the target pet is in a resting state based on the detection data includes: The feature parameters are input into a pre-trained activity detection model to determine whether the target pet is in the resting state.

6. The method as described in claim 5, characterized in that, Before inputting the feature parameters into the pre-trained activity detection model, the following steps are included: Receive training data from the inertial sensor; wherein the training data is measured by the inertial sensor when the target pet is in the preset position and when it leaves the preset position; Feature extraction is performed on the training data to obtain training feature parameters of the same type as the feature parameters; The activity detection model is trained using the aforementioned training feature parameters.

7. The method according to any one of claims 2 to 6, characterized in that, The detection data includes initial current data and multiple frames of initial historical data prior to the initial current data. Both the initial current data and each frame of the initial historical data have multiple data dimensions. The step of obtaining the current signal sequence based on the detection data includes: In each of the data dimensions, calculate the average value of the initial historical data for a first preset number of frames adjacent to the initial current data; The initial current data is subtracted from the average value in the corresponding data dimension to obtain the signal of the current frame, and then the signal of the current frame is combined with the signals of the multiple historical frames to form the current signal sequence.

8. The method according to any one of claims 2 to 6, characterized in that, The signal of the current frame and the signal of each historical frame have multiple data dimensions; the feature extraction of the current signal sequence to obtain feature parameters corresponding to the current frame includes: Feature extraction is performed on the current signal sequence to obtain multiple signal features related to the current frame, and the multiple signal features are combined into a feature vector.

9. The method as described in claim 8, characterized in that, The plurality of signal features includes the features of the current frame; the step of extracting features from the current signal sequence to obtain a plurality of signal features related to the current frame includes: Feature extraction is performed on the signal of the current frame to obtain the features of the current frame; wherein the features of the current frame include at least one of the mean, maximum value, minimum value, median or standard deviation of the signal of the current frame.

10. The method as described in claim 8, characterized in that, The step of extracting features from the current signal sequence to obtain multiple signal features related to the current frame includes: In each of the aforementioned data dimensions, the absolute value of the difference between the signal of the current frame and the signal of the historical frame adjacent to the current frame is calculated to obtain the differential signal; Feature extraction is performed on the differential signal to obtain the features of the differential signal; wherein, the plurality of signal features include the features of the differential signal, and the features of the differential signal include at least one of the mean, maximum value, minimum value, median or standard deviation of the differential signal.

11. The method as described in claim 8, characterized in that, The step of extracting features from the current signal sequence to obtain multiple signal features related to the current frame includes: The signal of the current frame is combined with the signal of the historical frame that is a second preset number of frames adjacent to the current frame to form a first sub-signal sequence; Feature extraction is performed on the first sub-signal sequence to obtain the features of the first sub-signal sequence; wherein, the plurality of signal features include the features of the first sub-signal sequence, and the features of the first sub-signal sequence include at least one of the standard deviation or covariance of the first sub-signal sequence in each data dimension.

12. The method as described in claim 8, characterized in that, The multiple signal features include confidence features; The step of extracting features from the current signal sequence to obtain multiple signal features related to the current frame includes: The signal of the current frame is combined with the signal of the historical frame that is a third preset number of frames adjacent to the current frame to form a second sub-signal sequence; The credibility of each data dimension is calculated based on the second sub-signal sequence; The maximum, mean, or median of the credibility of each of the data dimensions is used as the credibility feature.

13. The method as described in claim 12, characterized in that, The vital signs include respiratory parameters; the calculation of the reliability of each data dimension based on the second sub-signal sequence includes: In each of the data dimensions, the data of the second sub-signal sequence is smoothed to obtain a smoothed data sequence corresponding to the data dimension. Perform a Fast Fourier Transform on the smoothed data sequence to obtain the actual frequency domain signal; The actual maximum peak value is obtained by performing a full-domain peak search on the actual frequency domain signal; The target maximum peak value and the target second largest peak value are obtained by searching for peaks in the actual frequency domain signal within a preset interval; wherein, the preset interval corresponds to the respiratory parameter; Compare the actual maximum peak value with the target maximum peak value; If the actual maximum peak value is equal to the target maximum peak value, then the credibility of the corresponding data dimension is equal to the ratio of the target maximum peak value to the target second largest peak value; if the actual maximum peak value is not equal to the target maximum peak value, then the credibility of the corresponding data dimension is equal to the ratio of the actual maximum peak value to the target maximum peak value.

14. The method according to any one of claims 2 to 6, characterized in that, The vital signs include respiratory parameters; the current signal sequence has multiple data dimensions; determining the vital signs of the target pet based on the detection data includes: Within a first preset time window, the signal of the data dimension with the highest confidence in the current signal sequence is obtained to form a respiratory signal sequence; The respiratory signal sequence is bandpass filtered in a first preset frequency band to obtain the respiratory waveform signal of the target pet; wherein, the first preset frequency band is the pet's respiratory frequency band; and Based on the respiratory waveform signal, the respiratory parameters are determined; wherein the respiratory parameters include at least one of respiratory rate, respiratory rhythm, or respiratory pattern.

15. The method as described in claim 14, characterized in that, The vital signs include heart rate parameters; determining the vital signs of the target pet based on the detection data further includes: The heartbeat signal sequence is obtained by differentiating the respiratory signal sequence from the respiratory waveform signal. The heartbeat signal sequence is bandpass filtered in a second preset frequency band to obtain the heartbeat waveform signal of the target pet, where the second preset frequency band is the heartbeat frequency band of the pet; Based on the heartbeat waveform signal, the heartbeat parameters are determined; wherein the heartbeat parameters include at least one of heart rate or heart rate variability.

16. A device for detecting the vital signs of pets, characterized in that, include: The device body includes a preset location for accommodating the target pet; Inertial sensors are installed on the main body of the device; The detection component, communicatively connected to the inertial sensor, includes: At least one storage medium storing at least one instruction set for detecting the vital signs of the target pet; and At least one processor is communicatively connected to the at least one storage medium. When the detection device is running, the at least one processor reads the at least one instruction set and executes the method as described in any one of claims 1 to 15 according to the instructions of the at least one instruction set.

17. A system for detecting vital signs in pets, characterized in that, include: The pet vital signs detection device as described in claim 16; as well as The server is connected in communication with the detection device.

18. The detection system as described in claim 17, characterized in that, The detection system also includes: The terminal device is connected to the server.

Images