Intelligent cat litter box based on Internet of Things

Abstract

The invention relates to the technical field of pet equipment, in particular to an intelligent cat litter box based on the Internet of Things, which comprises a multi-pet recognition unit, a cat litter box control unit, a cat litter box control unit and a cat litter box control unit, the excrement detection unit is used for detecting organic components of excrement, the weight of the pet cat and the weight of the excrement; the abnormal behavior detection unit is used for detecting the defecation behavior characteristics of the pet cat; the data processing unit is used for receiving data of each unit, executing analysis logic and generating a health report of the pet cat; the data transceiving unit is used for interacting data with a user terminal through a wireless protocol; the power supply unit is used for receiving an external power supply, storing electric energy and supplying power to the smart cat litter basin; the user terminal is used for enabling an owner and a veterinarian to obtain health data of a pet cat online and adjust behaviors of the smart cat litter box; according to the method, a more comprehensive reference quantity is provided for health data of pet cats while different individuals in a multi-cat family are distinguished.

Description

Technical Field

[0001] The present invention relates to the technical field of pet equipment, and in particular to a smart cat litter box based on the Internet of Things. Background Art

[0002] With the prevalence of pet-keeping families in cities, pet health management has gradually become a core need for modern pet owners. Although traditional cat litter boxes can complete basic excrement collection, they lack the ability to deeply monitor the physiological status of pets. Existing smart cat litter boxes mostly focus on basic functions such as automatic cleaning and remote control, and have the following technical bottlenecks: Multi-cat identification defects: Market products rely on a single biometric feature (such as RFID collars), which are easily affected by environmental interference and cannot accurately distinguish individuals of similar breeds in multi-cat households; extensive health monitoring: Most devices only monitor excretion frequency or weight, and lack multi-dimensional correlation analysis of excrement composition and behavioral patterns; Insufficient data validity: The interference problem of cross-use of multiple cats has not been solved, resulting in low credibility of detection data. Based on this, there is an urgent need for an IoT cat litter box that integrates precise identification, multimodal detection and intelligent decision-making to solve the core pain points of pet health management in home scenarios.

[0003] Chinese patent application publication number CN118680093A discloses a cat litter box control device based on pet detection. The device includes a management and control module, a weight measurement module, a water content measurement module, and an intervention and interaction module. The weight measurement module obtains weight change data of the smart cat litter box. The management and control module analyzes the weight change data and obtains a health score for the pet cat based on the first, second, and third moments in the weight change data, as well as the first weight value corresponding to the first moment and the second weight value corresponding to the third moment. When the health score is less than a preset score threshold, the pet cat's excrement is analyzed for water content to obtain a water content percentage. A litter box control plan is generated based on the behavior duration, behavior frequency, and water content percentage, and the intervention and interaction module is controlled to interact with the user according to the litter box control plan.

[0004] It can be seen that the above-mentioned cat litter box control device based on pet detection has the problem of being unable to distinguish different individuals in a multi-cat household, and the cross-use of multiple cats leads to low credibility of the detection data. Summary of the Invention

[0005] To this end, the present invention provides a smart cat litter box based on the Internet of Things to overcome the problem in the prior art that different individuals in a multi-cat household cannot be distinguished, and the cross-use of multiple cats leads to low credibility of detection data.

[0006] To achieve the above object, the present invention provides a smart cat litter box based on the Internet of Things, comprising:

[0007] A multi-pet identification unit is installed on one side of the cat litter box entrance to collect the biological characteristics of animals and distinguish different individuals;

[0008] The excrement detection unit includes a component detection module disposed on the top of the cat litter box to detect the organic components of the excrement and a weight detection module disposed on the bottom of the cat litter box to detect the weight of the pet cat and the weight of the excrement;

[0009] An abnormal behavior detection unit is provided inside the bowl of the smart cat litter box to detect the toileting behavior characteristics of the pet cat;

[0010] a data processing unit connected to the multi-pet identification unit, the excrement detection unit, the abnormal behavior detection unit, the power supply unit, and the data transceiver unit, for receiving data from each unit, executing analysis logic, and generating a health report for the pet cat;

[0011] A data transceiver unit, connected to the data processing unit and the power supply unit, for exchanging data with the user terminal via a wireless protocol;

[0012] a power supply unit connected to the multi-pet identification unit, the excrement detection unit, the abnormal behavior detection unit, the data processing unit, and the data transceiver unit, and disposed on a side away from the entrance of the cat litter box, for receiving an external power supply and storing electrical energy to supply power to the smart cat litter box;

[0013] The user terminal includes an owner terminal and a veterinarian terminal that are connected to each other to exchange information. The user terminal interacts with the data transceiver unit through a wireless protocol, so that the owner and the veterinarian can obtain the health data of the pet cat online and adjust the behavior of the smart litter box.

[0014] Furthermore, the multi-pet identification unit includes a camera module arranged above the entrance of the smart cat litter box for photographing the face of the pet cat, a movable door arranged at the entrance of the smart cat litter box for judging and restricting the entry and exit of the pet cat, and a bioelectric signal acquisition module arranged outside the entrance of the smart cat litter box for collecting the mechanical characteristic parameters and foot epidermal impedance parameters of the pet cat.

[0015] Furthermore, the component detection module includes:

[0016] The excrement locating mechanism includes a plurality of infrared thermal imaging sensors for detecting changes in the surface temperature of the cat litter to locate the location of the excrement;

[0017] An anti-interference sampling mechanism includes a planar movable guide rail and a liftable shielding cover disposed on the planar movable guide rail for forming a physically isolated space in the excrement location area;

[0018] The multi-spectral scanning mechanism is arranged inside the liftable shielding cover and is used for performing non-contact scanning on the excrement of the pet cat inside the liftable shielding cover.

[0019] Furthermore, the abnormal behavior detection unit includes a three-axis acceleration sensor arranged in the central area of ​​the bottom of the cat litter box for detecting the posture change frequency of the pet cat when going to the toilet, and a sound sensor arranged on the inner side wall of the cat litter box for detecting the strength of the pet cat's sand scraping.

[0020] Furthermore, when the movable door recognizes that a pet cat has entered or exited once, the data processing unit determines that an excretion event has been completed, generates a health report with a timestamp, and controls the data transceiver unit to send the health report to the owner's terminal and store it; the health report includes: the pet cat's individual identity information, bioelectrical signal characteristics, toileting behavior characteristics, and excrement detection data;

[0021] The bioelectrical signal characteristics include: foot mechanical characteristic parameters and foot epidermal impedance characteristic parameters;

[0022] The toileting behavior characteristics include: sand-scraping intensity, posture change frequency and toileting duration;

[0023] The excrement detection data includes: excrement weight data, excretion frequency data and excrement organic component data.

[0024] Furthermore, when the multi-pet recognition unit recognizes that a pet cat is approaching the entrance of the cat litter box, the data processing unit controls the camera module to capture image data of the pet cat, and calculates a first recognition confidence based on nose print data in the image data to preliminarily determine the individual pet cat;

[0025] When the movable door recognizes that the pet cat has entered the litter box, it is determined that an excretion event has begun, and a second confidence level of recognition is calculated based on the biomechanical characteristics of the pet cat's foot and the foot epidermal impedance parameters acquired by the bioelectrical signal acquisition module;

[0026] The pet cat individual is determined according to the first recognition confidence and the second recognition confidence.

[0027] Furthermore, when the foot mechanical characteristic parameter or the foot epidermal impedance characteristic parameter of the pet cat individual exceeds a preset range, the data processing unit marks the health report of the current pet cat individual, controls the data transceiver unit to send the health report to the owner terminal while reminding the owner, and applies to the owner to call the historical health report of the current pet cat individual;

[0028] When the application is approved, the current health report and the historical health report are sent to the veterinary terminal and a reminder is issued.

[0029] Furthermore, the data processing unit calculates an abnormality index of the current pet cat's excretion behavior based on the current pet cat's sand-scraping strength, posture change frequency, and toileting duration detected by the abnormal behavior detection unit;

[0030] When the abnormality index is greater than a first threshold, the health report of the current pet cat individual is marked, and the data transceiver unit is controlled to send the health report to the owner terminal while reminding;

[0031] When the abnormality index is greater than a second threshold, the health report of the current pet cat individual is marked, the data transceiver unit is controlled to send the health report to the owner terminal while reminding the owner, and the owner is requested to call the historical health report of the current pet cat individual;

[0032] When the application is approved, the current health report and the historical health report are sent to the veterinary terminal and a reminder is issued.

[0033] Furthermore, the defecation frequency data is the total time interval between several defecations of a pet cat. When the total time interval is less than or equal to the defecation frequency threshold, the most recent health report of the current pet cat is marked, the data transceiver unit is controlled to send the health report to the owner terminal while reminding the owner, and the owner is requested to call the historical health report of the current pet cat.

[0034] When the application is approved, the current health report and the historical health report are sent to the veterinary terminal and a reminder is issued.

[0035] The excretion frequency threshold is determined based on the basic excretion frequency threshold and historical data of different pet cat individuals.

[0036] Furthermore, when the movable door in the multi-pet recognition unit recognizes that a single pet has entered, the entry time is recorded, and based on a preset time period set for the corresponding pet, the opening and closing angle of the movable door from the outside to the inside is locked within the time period to prevent other pets from entering;

[0037] The preset duration is determined based on the historical usage duration of a single pet or a pre-set duration.

[0038] Compared with the existing technology, the beneficial effect of the present invention is that, through multi-sensor fusion and Internet of Things technology, the present invention constructs a closed-loop health management system covering biometrics, excretion analysis, and behavior monitoring, which significantly improves the data accuracy and timeliness of health warnings in multi-cat household scenarios.

[0039] Furthermore, the present invention achieves high-precision binding of excretion data and individual identity through the collaborative architecture of multiple pet identification units and excrement detection units, avoiding confusion of multi-cat household data. At the same time, through the dual-modal recognition of bioelectric signals and visual features and the dual-reset confidence cross-validation mechanism, it can effectively distinguish individuals with similar hair colors, greatly reducing the risk of health data dislocation caused by misjudgment.

[0040] Furthermore, the present invention uses the way a pet cat's feet touch the ground and the characteristic resistance value of the foot's stratum corneum as characteristic signs for pet cat identification. While greatly increasing the accuracy of pet cat identification in multi-cat households, it also provides a more comprehensive reference for pet cat health data. Compared with traditional observation methods, chronic diseases such as arthritis and skin diseases in pet cats can be discovered in advance.

[0041] Furthermore, the present invention uses non-contact multispectral scanning combined with a physical isolation cover to ensure that the detection of fecal components is not interfered with by the feces of other individuals, while avoiding the risk of cross-contamination of contact probes.

[0042] Furthermore, the present invention can identify abnormal frequency of changes in excretion posture (such as frequent adjustments in body position caused by urination pain) and abnormal sand-shaving force (such as fatigue characteristics caused by kidney disease) through multi-source behavior capture using a three-axis acceleration sensor and a sound sensor.

[0043] Furthermore, the present invention achieves early screening of urinary diseases and digestive system problems through a graded early warning system based on abnormality indexes, by dynamically correlating excrement weight, composition and behavioral data;

[0044] Furthermore, the present invention significantly improves the detection sensitivity of metabolic diseases such as diabetes and hyperthyroidism through the analysis logic that links the excretion frequency threshold with historical data.

[0045] Furthermore, the present invention effectively prevents detection conflicts caused by the simultaneous use of multiple cats through a time locking strategy, thereby ensuring the integrity and traceability of single excretion data. BRIEF DESCRIPTION OF THE DRAWINGS

[0046] Figure 1 This is a schematic structural diagram of the Internet of Things-based smart cat litter box of the present invention;

[0047] Figure 2 This is a connection block diagram of the smart cat litter box based on the Internet of Things of the present invention;

[0048] Figure 3 A schematic diagram of the data structure of a health report of a smart cat litter box based on the Internet of Things of the present invention;

[0049] In the figure, 11-CMOS camera; 12-movable door; 131-piezoresistive film sensor; 132-flexible capacitive sensor; 211-infrared thermal imaging sensor; 2121-planar movable guide rail; 2122-hydraulic telescopic rod; 2123-bell-shaped black shielding cover; 2131-LED array; 2132-CMOS spectral sensor; 22-resistive strain sensor; 31-infrared thermal imaging sensor; 32-wideband piezoelectric sound sensor; 4-data processing unit; 5-data transceiver unit; 6-power supply unit. DETAILED DESCRIPTION

[0050] In order to make the objects and advantages of the present invention more clearly understood, the present invention is further described below in conjunction with embodiments; it should be understood that the specific embodiments described herein are merely used to explain the present invention and are not intended to limit the present invention.

[0051] The preferred embodiments of the present invention are described below with reference to the accompanying drawings. It should be understood by those skilled in the art that these embodiments are only used to explain the technical principles of the present invention and are not intended to limit the scope of protection of the present invention.

[0052] It should be noted that, in the description of the present invention, terms such as "up", "down", "left", "right", "inside", and "outside" indicating directions or positional relationships are based on the directions or positional relationships shown in the accompanying drawings. This is only for the convenience of description and does not indicate or imply that the device or element must have a specific orientation, be constructed and operated in a specific orientation. Therefore, it cannot be understood as a limitation on the present invention.

[0053] Furthermore, it should be noted that, in the description of the present invention, unless otherwise expressly specified or limited, the terms "mounted," "connected," and "connected" should be understood in a broad sense. For example, they may refer to fixed connections, detachable connections, or integral connections; mechanical connections or electrical connections; direct connections or indirect connections through an intermediate medium; and internal communication between two components. Those skilled in the art will understand the specific meanings of the above terms in the present invention based on specific circumstances.

[0054] Specific embodiments are given below in conjunction with the accompanying drawings. The specific embodiments are only used to describe the technical solutions of the present invention in detail and are not intended to limit the scope of protection of the present application.

[0055] The embodiment of the present invention provides a smart cat litter box based on the Internet of Things, see Figures 1 to 3 As shown, it is a schematic diagram of the structure of the smart cat litter box based on the Internet of Things of the present invention, a connection block diagram of the smart cat litter box based on the Internet of Things of the present invention, and a schematic diagram of the data structure of the health report of the smart cat litter box based on the Internet of Things of the present invention; including:

[0056] A multi-pet identification unit, located at the entrance of the litter box, collects the biological characteristics of pet cats and distinguishes between different individuals;

[0057] Specifically, the multi-pet identification unit includes a camera module arranged above the entrance of the smart cat litter box for photographing the face of the pet cat, a movable door arranged at the entrance of the smart cat litter box for judging and restricting the entry and exit of the pet cat, and a bioelectric signal acquisition module arranged outside the entrance of the smart cat litter box for collecting the mechanical characteristic parameters and epidermal impedance parameters of the pet cat's feet.

[0058] In the implementation, the camera module is a CMOS camera 11, preferably, its specifications are global shutter, resolution 1280×720, and frame rate 30fps; the bioelectric signal acquisition module is a flat pad composed of a lower layer of piezoresistive film sensor 131 and an upper layer of flexible capacitive electrode array 132, preferably, the specifications of the piezoresistive film sensor are 16×16 dot matrix, and the resolution is 1cm 2 / point; preferably, the flexible capacitive electrode array 132 is a dot matrix electrode made of conductive silicone, with a specification of 8×8 dot matrix, and the electrode surface is covered with a hydrophobic coating.

[0059] It is understandable that the above preferred scheme is determined through experiments and historical data, which can better balance the power consumption and the accuracy requirements of multi-pet recognition. Those skilled in the art can replace the form and specifications of the above modules while achieving the same effect, and will not go into details here.

[0060] The excrement detection unit includes a component detection module disposed on the top of the cat litter box to detect the organic components of the excrement and a weight detection module disposed on the bottom of the cat litter box to detect the weight of the pet cat and the weight of the excrement;

[0061] Specifically, the component detection module includes:

[0062] The excrement locating mechanism includes a plurality of infrared thermal imaging sensors 211 for detecting changes in the surface temperature of the cat litter to locate the location of the excrement;

[0063] In implementation, the infrared thermal imaging sensor 211 is an uncooled micro-thermometry infrared sensor. Preferably, two uncooled micro-thermometry infrared sensors with a thermal sensitivity of less than 80 mK are used.

[0064] The anti-interference sampling mechanism includes a planar movable guide rail 2121 and a liftable shielding cover disposed on the planar movable guide rail 2121 to form a physically isolated space in the excrement location area;

[0065] In implementation, the liftable shielding cover includes a hydraulic telescopic rod 2122 and a bell-shaped black shielding cover 2123 arranged at the end of the hydraulic telescopic rod 2122. Preferably, the light intensity attenuation of the bell-shaped black shielding cover 2123 is greater than or equal to 90% to isolate the interference of ambient light and other excrement spectra.

[0066] The multi-spectral scanning mechanism is arranged inside the liftable shielding cover and is used for performing non-contact scanning on the excrement of the pet cat inside the liftable shielding cover.

[0067] The multi-spectral scanning mechanism includes an LED array 2131 and a CMOS spectrum sensor 2132. Preferably, the LED array 2131 is a near-infrared LED array, covering the 850nm to 1650nm band (containing characteristic absorption peaks of water, urea, and creatinine) with a wavelength accuracy of ±2nm, so as to detect the content of creatinine, urea, and water in the excreta;

[0068] The weight detection module is a resistance strain type weighing sensor 22, and preferably, its measuring range is 0.5kg to 10kg.

[0069] It is understandable that the above preferred scheme is determined through experiments and historical data, which can better balance the power consumption and the accuracy requirements of pet excrement composition and weight detection. Those skilled in the art can replace the form and specifications of the above modules while achieving the same effect, and will not go into details here.

[0070] An abnormal behavior detection unit is provided inside the bowl of the smart cat litter box to detect the toileting behavior characteristics of the pet cat;

[0071] Specifically, the abnormal behavior detection unit includes a three-axis acceleration sensor 31 arranged in the center area of ​​the bottom of the cat litter box for detecting the posture change frequency of the pet cat when going to the toilet, and a sound sensor arranged on the inner side wall of the cat litter box for detecting the strength of the pet cat's sand scraping.

[0072] In implementation, the three-axis acceleration sensor 31 preferably has a range of ±8g and a sampling rate of 50Hz; the sound sensor is a wide-band piezoelectric sound sensor 32, preferably a wide-band piezoelectric sound sensor 32 with a frequency response of 20Hz to 20kHz.

[0073] It is understandable that the above-mentioned preferred scheme is determined through experiments and historical data, which can better balance the power consumption and accuracy requirements of toilet behavior feature detection. Those skilled in the art can replace the form and specifications of the above-mentioned modules while obtaining the same effect, and will not go into details here.

[0074] a data processing unit 4 connected to the multi-pet identification unit, the excrement detection unit, the abnormal behavior detection unit, the power supply unit 6, and the data transceiver unit 5, for receiving data from each unit, executing analysis logic, and generating a health report for the pet cat;

[0075] a data transceiver unit 5 connected to the data processing unit 4 and the power supply unit 6 for exchanging data with the user terminal via a wireless protocol;

[0076] It is understandable that the data interaction function of the data transceiver unit 5 can be applied to any implementation form in the prior art, and will not be described in detail here.

[0077] a power supply unit 6, connected to the multi-pet identification unit, the excrement detection unit, the abnormal behavior detection unit, the data processing unit 4, and the data transceiver unit 5, and disposed on a side away from the entrance of the cat litter box, for receiving an external power supply and storing electrical energy to power the smart cat litter box;

[0078] The user terminal includes an owner terminal and a veterinarian terminal that can exchange information with each other, and exchanges data with the data transceiver unit through a wireless protocol, so that the owner and the veterinarian can obtain the health data of the pet cat online and adjust the behavior of the smart cat litter box.

[0079] Specifically, when the movable door 12 identifies a pet cat entering or exiting once, the data processing unit 4 determines that an excretion event has been completed, generates a health report with a timestamp, and controls the data transceiver unit 5 to send the health report to the owner's terminal and store it; the health report includes: the pet cat's individual identity information, bioelectrical signal characteristics, toileting behavior characteristics, and excrement detection data;

[0080] The bioelectrical signal characteristics include: foot mechanical characteristic parameters and foot epidermal impedance characteristic parameters;

[0081] The toileting behavior characteristics include: sand-scraping intensity, posture change frequency and toileting duration;

[0082] The excrement detection data includes: excrement weight data, excretion frequency data and excrement organic component data.

[0083] The excrement detection data includes excrement weight data and excrement organic component data.

[0084] In implementation, the excreta organic component data are the urea / creatinine ratio and water content of the excreta.

[0085] It is understandable that the urea / creatinine ratio and water content are core indicators of renal function, which can be measured more accurately in a non-laboratory environment using spectral sensors and are suitable as targets for pathological examination of pet cat excreta in a home environment.

[0086] Specifically, when the multi-pet recognition unit recognizes that a pet cat is approaching the entrance of the cat litter box, the data processing unit 4 controls the image data of the pet cat captured by the camera module, and calculates the first recognition confidence based on the nose print data in the image data to preliminarily determine the individual pet cat.

[0087] When the movable door 12 recognizes that the pet cat has entered the litter box, it is determined that the excretion event has started, and the bioelectric signal acquisition module collects the biomechanical characteristics and foot epidermal impedance parameters of the pet cat's foot, and calculates the second confidence level of recognition based on the foot mechanical characteristic parameters and foot epidermal impedance characteristic parameters;

[0088] The pet cat individual is determined according to the first recognition confidence and the second recognition confidence.

[0089] In implementation, when the sum of the first recognition confidence and the second recognition confidence is less than a confidence threshold, the excretion detection of the individual pet cat is marked as uncertain, and an encrypted video clip of the pet cat's image data is attached to the excrement detection data of this excretion event to assist the owner and veterinarian in making a judgment;

[0090] When the sum of the first recognition confidence and the second recognition confidence is greater than or equal to the confidence threshold, it is confirmed that the individual identification of the pet cat is successful, and the individual identity information of the pet cat is associated with the health report.

[0091] In practice, before the present invention is put into use, the owner is required to take a picture of the face of each pet cat using a CMOS image sensor, and collect a nose print feature point set T = {t1, t2, ..., t n}, and used as the nose print feature in the pre-stored template of each pet cat, where n is the total number of nose print feature points.

[0092] In implementation, a 2-3 second video of the pet cat is continuously captured by a CMOS image sensor, and the first recognition confidence C1 is calculated, including:

[0093] Extract the nose print feature point set S = {s1, s2, ..., s n}; Call the nose print feature point set T = {t1, t2, ..., t n}, and calculate the cosine similarity S1 between the nose print feature point set S and the nose print feature point set T in each pre-stored template, where n≥50;

[0094] The calculation formula of the cosine similarity S1 is specifically:

[0095]

[0096] Then, the image blurriness of the video is calculated and recorded as D. In the implementation, the image blurriness D is obtained based on the grayscale variance algorithm. Specifically, after converting the image to a grayscale image, the grayscale average value of all pixels in the image is used as a reference, the grayscale value of each pixel is subtracted and the square sum is calculated, and then normalized by the total number of pixels. It represents the average degree of grayscale change in the image. The greater the average degree of grayscale change, the clearer the image, and the smaller the average degree of grayscale change, the blurrier the image. This algorithm is prior art and will not be described in detail here.

[0097] The first recognition confidence C1 is calculated and its calculation formula is specifically as follows:

[0098]

[0099] Wherein, α is the nose print weight coefficient, preferably, α=0.8; β is the image blur compensation coefficient, preferably, β=0.2; D min is the image blur threshold, preferably, 500;

[0100] Take the pet cat individual in the pre-stored template corresponding to the maximum value of all the first confidence levels C1 as the initially determined pet cat individual;

[0101] It can be understood that the cosine similarity is used to quantify the spatial distribution consistency of the current pet cat's individual nose print and the template to avoid the influence of lighting changes. The closer its value is to 1, the more similar it is. At the same time, the image blur correction term is introduced. The larger its value is, the clearer the image is. min =500 is taken as the minimum acceptable blur level of the image. When the image blur is greater than 500, the first recognition confidence is increased. When the image blur is less than 500, the first recognition confidence is decreased.

[0102] When the movable door 12 is opened inwards at an angle greater than 80°, it is determined that the pet cat has entered the litter box and the excretion event begins;

[0103] Specifically, calculating the second recognition confidence C2 includes:

[0104] The biomechanical characteristics of the foot are as follows:

[0105] The piezoresistive film sensor 131 records the contact area A of the pet cat's paw pad. i (Unit: cm 2) matrix data that changes with time, with a sampling rate of 2kHz for 3 seconds; and calculate the normalized pressure distribution P i , and its calculation formula is as follows:

[0106] i is a corresponding 16×16 single dot matrix of the piezoresistive thin film sensor 131 .

[0107] According to the normalized pressure distribution P i Calculate the pressure distribution entropy value H p , and its calculation formula is:

[0108] The unit is bit;

[0109] Specifically, the foot epidermal impedance parameters are as follows:

[0110] When the pet cat's paw pad contacts the flexible capacitive electrode array 132 , a swept frequency AC signal is applied. Preferably, the swept frequency AC signal is 10 Hz to 100 Hz, and a typical excitation voltage is 5 mV to 25 mV.

[0111] Record the impedance amplitude |Z(f)| and phase angle θ(f) curves as they change with frequency; extract the frequency corresponding to the minimum phase angle θ and record it as the foot epidermal impedance f p ,Right now:

[0112] The unit is kHz;

[0113] The bioelectric characteristic baseline value in the pre-stored template of the initially determined pet cat individual is called, including:

[0114]

[0115] Among them, μ H H p The average value of σ H H p The standard deviation of

[0116] In practice, before the present invention is put into use, the owner is required to guide each pet cat to enter the litter box normally at least 5 times in a row, and collect 5 to 10 seconds of foot pressure data and impedance parameter data each time to calculate H p The average value μ H and standard deviation σ H , and f p The maximum and minimum values ​​are used as the bioelectric characteristic baseline values ​​in the pre-stored template of the pet cat individual.

[0117] The pressure distribution entropy deviation ΔH is calculated, and its calculation formula is as follows:

[0118]

[0119] The second recognition confidence C2 is calculated and the calculation formula is as follows:

[0120]

[0121] Among them, when f p lie in When , f = 1; otherwise, f = 0;

[0122] The confidence threshold is preferably between 0.8 and 0.95.

[0123] Specifically, when the foot mechanical characteristic parameter or the foot epidermal impedance characteristic parameter of the pet cat individual exceeds a preset range, the data processing unit 4 marks the health report of the current pet cat individual, controls the data transceiver unit 5 to send the health report to the owner terminal while reminding the owner, and applies to the owner to call the historical health report of the current pet cat individual;

[0124] When the application is approved, the current health report and the historical health report are sent to the veterinary terminal and a reminder is issued.

[0125] In practice, for the biomechanical characteristics of the foot, the characterization parameter is the pressure distribution entropy value H p , preferably, its normal value is H p =2.1±0.3; For the foot epidermal impedance parameter, its characterization parameter is the foot epidermal impedance f p , preferably, its normal value is f p =12.5±3kHz.

[0126] It is understandable that for healthy adult cats, H p =2.1±0.4, that is, the pressure distribution entropy is low, indicating that the pressure distribution is regular; for arthritic cats, H p >2.5, that is, the pressure distribution entropy is high, indicating that pain causes uneven ground contact; for adult healthy cats, f p =12.5±3kHz, this range is determined by the dielectric properties of the stratum corneum of the pet cat's epidermis; for elderly cats or cats with dry skin, f p <8kHz, indicating that the thickening of the stratum corneum causes a low-frequency shift.

[0127] The present invention uses the way a pet cat's feet touch the ground and the characteristic resistance value of the foot's stratum corneum as characteristic markers for pet cat identification. While greatly increasing the accuracy of pet cat identification in multi-cat households, it also provides a more comprehensive reference for pet cat health data. Compared with traditional observation methods, chronic diseases such as arthritis and skin diseases in pet cats can be discovered in advance.

[0128] Specifically, the data processing unit 4 calculates the abnormal index of the current pet cat's excretion behavior based on the current pet cat's sand-scraping strength, posture change frequency and toileting time detected by the abnormal behavior detection unit;

[0129] When the abnormality index is greater than a first threshold, the health report of the current pet cat individual is marked, and the data transceiver unit 5 is controlled to send the health report to the owner terminal while reminding;

[0130] When the abnormality index is greater than a second threshold, the health report of the current pet cat individual is marked, the data transceiver unit 5 is controlled to send the health report to the owner terminal while reminding the owner, and the owner is requested to call the historical health report of the current pet cat individual;

[0131] When the application is approved, the current health report and the historical health report are sent to the veterinary terminal and a reminder is issued.

[0132] In practice, the planing force L is obtained by A-weighted filtering through the broadband piezoelectric sound sensor 32, and the unit is (dB);

[0133] The sand-shaving force L is dimensionlessly processed, specifically:

[0134] L'=(LL min ) / (L max -L min );

[0135] Among them, preferably, L min =40dB, which is the reference value for a pet cat when it is quietly scraping the sand; L max =75dB, which is the reference value for cats anxiously scraping litter;

[0136] The posture change frequency F is calculated by the three-axis acceleration sensor 31 and the unit is (times / minute);

[0137] Specifically, the three-axis acceleration sensor 31 is used to collect the X / Y / Z acceleration (unit: gravitational acceleration g) of the bottom of the litter box of the pet cat in real time during the defecation process and calculate the composite acceleration: the variance of the acceleration change amplitude within the sliding window is calculated based on the composite acceleration. Preferably, the sliding window time length is 1 second, and there are 50 sampling points in a single sliding window;

[0138] When the variance in three consecutive sliding windows exceeds 0.4g 2 When , it is determined that a posture change occurs (such as standing up, lying down, turning around, etc.);

[0139] The attitude change frequency is dimensionlessly processed, specifically:

[0140] F'=F / F max ,

[0141] Among them, F max =4 times / minute, which is the maximum reference value of posture change frequency of a healthy cat;

[0142] The toilet duration M is the duration of the excretion event detected by the movable door 12, in seconds;

[0143] The toilet time M is dimensionlessly processed as follows:

[0144] M'=|MM avg | / M avg ,

[0145] Among them, M avg Take the average of the cat's toileting time over the past three days;

[0146] The calculation formula of abnormal index E is as follows:

[0147] E=w1×L'+w2×F'+w3×M';

[0148] Among them, w1, w2 and w3 are the weight coefficients of sanding force, posture change frequency and toilet duration respectively; preferably, w1=0.4, w2=0.3, w3=0.3.

[0149] In practice, the first threshold is 0.6 and the second threshold is 1.2. They can be obtained by obtaining a limited number of pet data by taking the average or maximum value of the abnormality index of pet cats with normal physical signs, which will not be described in detail here.

[0150] It is understandable that the intensity of sand scraping is negatively correlated with the degree of pain in kidney disease, abnormal posture frequency can reflect the body position adjustment caused by urinary stones, and the duration of toileting is related to urinary diseases and digestive system problems.

[0151] Specifically, the defecation frequency data is the total time interval between several defecations of a pet cat. When the total time interval is less than or equal to the defecation frequency threshold, the most recent health report of the current pet cat is marked, and the data transceiver unit 5 is controlled to send the health report to the owner terminal while reminding the owner, and requesting the owner to call the historical health report of the current pet cat individual;

[0152] When the application is approved, the current health report and the historical health report are sent to the veterinary terminal and a reminder is issued.

[0153] Specifically, when the movable door 21 in the multi-pet recognition unit recognizes that a single pet has entered, the entry time is recorded, and based on the preset time period set for the corresponding pet, the opening and closing angle of the movable door 21 from the outside to the inside is locked during the period to prevent other pets from entering;

[0154] The preset duration is determined based on the historical usage duration of a single pet or a pre-set duration.

[0155] During implementation, when the movable door 21 in the multi-pet identification unit recognizes that pet cat A enters, the entry time T1 is recorded, and the opening and closing angle of the movable door 21 from the outside to the inside is locked within the time period [T1, T1+t); at the same time, when pet cat B attempts to enter, if there is an uncompleted excretion event in the current time period, the identification process of pet cat B is delayed until the detection of the current excretion event is completed, where t is the preset time length.

[0156] The present invention may further include an energy collection unit, which is disposed in the basin frame of the cat litter box and connected to the power supply unit 6, for converting mechanical vibration energy into electrical energy and inputting the energy into the power supply unit 6 for storage;

[0157] In practice, when the energy harvesting unit converts mechanical vibration energy into electrical energy, it includes a vibration energy harvesting mechanism and a vibration amplification mechanism. The vibration energy harvesting mechanism is composed of alternating layers of piezoelectric fiber bundles and magnetostrictive sheets, which are radially arranged within the basin support frame. The vibration amplification mechanism includes a spring array and a mass tuning system, which couples weak vibration energy from different parts of the basin to the vibration energy harvesting mechanism and amplifies it.

[0158] The vibration energy capture mechanism collects the multimodal mechanical vibrations generated when the pet enters and exits, and after the energy density is increased by the vibration amplification mechanism, it is converted into direct current and transmitted to the power supply unit 6 to store the electrical energy.

[0159] It is understandable that the above-mentioned vibration energy capture mechanism and vibration amplification mechanism are existing technologies and will not be described in detail here.

[0160] In implementation, when the power supply unit 6 is connected to an external power source, the energy collection unit is stopped, and the multi-pet identification unit, the excrement detection unit, and the abnormal behavior detection unit are controlled to work at rated power and open all functions;

[0161] When the power supply unit 6 is not connected to an external power source, turning on the energy collection unit;

[0162] If the remaining power is greater than 70%, the multi-pet identification unit, the excrement detection unit, and the abnormal behavior detection unit are controlled to operate at rated power and all functions are enabled;

[0163] If 40% ≤ remaining power ≤ 70%, the remaining power is allocated according to priority, where the priority is: excretion detection > behavior monitoring > identity recognition;

[0164] If the remaining energy is less than 40%, the component detection module and the abnormal behavior detection unit are turned off, and a power reminder is sent to the host terminal via the data transceiver unit.

[0165] Thus far, the technical solutions of the present invention have been described in conjunction with the preferred embodiments shown in the accompanying drawings. However, it is readily understood by those skilled in the art that the scope of protection of the present invention is obviously not limited to these specific embodiments. Without departing from the principles of the present invention, those skilled in the art may make equivalent changes or substitutions to the relevant technical features, and the technical solutions after such changes or substitutions will fall within the scope of protection of the present invention.

Claims

1. A smart cat litter box based on the Internet of Things, characterized in that: include: A multi-pet identification unit is installed on one side of the cat litter box entrance to collect the biological characteristics of animals and distinguish different individuals; The excrement detection unit includes a component detection module disposed on the top of the cat litter box to detect the organic components of the excrement and a weight detection module disposed on the bottom of the cat litter box to detect the weight of the pet cat and the weight of the excrement; An abnormal behavior detection unit is provided inside the bowl of the smart cat litter box to detect the toileting behavior characteristics of the pet cat; a data processing unit connected to the multi-pet identification unit, the excrement detection unit, the abnormal behavior detection unit, the power supply unit, and the data transceiver unit, for receiving data from each unit, executing analysis logic, and generating a health report for the pet cat; A data transceiver unit, connected to the data processing unit and the power supply unit, for exchanging data with the user terminal via a wireless protocol; a power supply unit connected to the multi-pet identification unit, the excrement detection unit, the abnormal behavior detection unit, the data processing unit, and the data transceiver unit, and disposed on a side away from the entrance of the cat litter box, for receiving an external power supply and storing electrical energy to supply power to the smart cat litter box; The user terminal includes an owner terminal and a veterinarian terminal that are connected to each other to exchange information. The user terminal interacts with the data transceiver unit through a wireless protocol, so that the owner and the veterinarian can obtain the health data of the pet cat online and adjust the behavior of the smart litter box.

2. The smart cat litter box based on the Internet of Things according to claim 1, characterized in that, The multi-pet identification unit includes a camera module arranged above the entrance of the smart cat litter box for photographing the face of the pet cat, a movable door arranged at the entrance of the smart cat litter box for judging and restricting the entry and exit of the pet cat, and a bioelectric signal acquisition module arranged outside the entrance of the smart cat litter box for collecting the mechanical characteristic parameters and foot epidermal impedance parameters of the pet cat.

3. The smart cat litter box based on the Internet of Things according to claim 1, characterized in that, The component detection module includes: The excrement locating mechanism includes a plurality of infrared thermal imaging sensors for detecting changes in the surface temperature of the cat litter to locate the location of the excrement; An anti-interference sampling mechanism includes a planar movable guide rail and a liftable shielding cover disposed on the planar movable guide rail for forming a physically isolated space in the excrement location area; The multi-spectral scanning mechanism is arranged inside the liftable shielding cover and is used for performing non-contact scanning on the excrement of the pet cat inside the liftable shielding cover.

4. The smart cat litter box based on the Internet of Things according to claim 1, characterized in that The abnormal behavior detection unit includes a three-axis acceleration sensor arranged in the center area of ​​the bottom of the cat litter box for detecting the posture change frequency of the pet cat when going to the toilet, and a sound sensor arranged on the inner side wall of the cat litter box for detecting the intensity of the pet cat's sand scraping.

5. The smart cat litter box based on the Internet of Things according to claim 2, characterized in that: When the movable door recognizes that a pet cat has entered or exited once, the data processing unit determines that an excretion event has been completed, generates a health report with a timestamp, and controls the data transceiver unit to send the health report to the owner's terminal and store it; the health report includes: the pet cat's individual identity information, bioelectrical signal characteristics, toileting behavior characteristics, and excrement detection data; The bioelectrical signal characteristics include foot mechanical characteristic parameters and foot epidermal impedance characteristic parameters; The toileting behavior characteristics include sand-scraping intensity, posture change frequency, and toileting duration; The excrement detection data includes: excrement weight data, excretion frequency data and excrement organic component data.

6. The smart cat litter box based on the Internet of Things according to claim 2, characterized in that: When the multi-pet recognition unit recognizes that a pet cat is approaching the entrance of the cat litter box, the data processing unit controls the camera module to capture image data of the pet cat, and calculates a first recognition confidence level based on nose print data in the image data to preliminarily determine the individual pet cat; When the movable door recognizes that the pet cat has entered the litter box, it is determined that an excretion event has begun, and a second confidence level of recognition is calculated based on the biomechanical characteristics of the pet cat's foot and the foot epidermal impedance parameters acquired by the bioelectrical signal acquisition module; The pet cat individual is determined according to the first recognition confidence and the second recognition confidence.

7. The smart cat litter box based on the Internet of Things according to claim 6, characterized in that: The data processing unit, when the foot mechanical characteristic parameter or the foot epidermal impedance characteristic parameter of the pet cat individual exceeds a preset range, marks the health report of the current pet cat individual, controls the data transceiver unit to send the health report to the owner terminal while reminding the owner, and applies to the owner to call the historical health report of the current pet cat individual; When the application is approved, the current health report and the historical health report are sent to the veterinary terminal and a reminder is issued.

8. The smart cat litter box based on the Internet of Things according to claim 5, characterized in that: The data processing unit calculates an abnormality index of the current pet cat's excretion behavior based on the current pet cat's sand-scraping strength, posture change frequency, and toileting time detected by the abnormal behavior detection unit; When the abnormality index is greater than a first threshold, the health report of the current pet cat individual is marked, and the data transceiver unit is controlled to send the health report to the owner terminal while reminding; When the abnormality index is greater than a second threshold, the health report of the current pet cat individual is marked, the data transceiver unit is controlled to send the health report to the owner terminal while reminding the owner, and the owner is requested to call the historical health report of the current pet cat individual; When the application is approved, the current health report and the historical health report are sent to the veterinary terminal and a reminder is issued.

9. The smart cat litter box based on the Internet of Things according to claim 5, characterized in that: The defecation frequency data is the total time interval between several defecations of a pet cat. When the total time interval is less than or equal to the defecation frequency threshold, the most recent health report of the current pet cat is marked, the data transceiver unit is controlled to send the health report to the owner terminal while reminding the owner, and the owner is requested to call the historical health report of the current pet cat; When the application is approved, the current health report and the historical health report are sent to the veterinary terminal and a reminder is issued. The excretion frequency threshold is determined based on the basic excretion frequency threshold and historical data of different pet cat individuals.

10. The smart cat litter box based on the Internet of Things according to claim 2, characterized in that: When the movable door in the multi-pet recognition unit recognizes that a single pet has entered, the entry time is recorded, and based on a preset time period set for the corresponding pet, the opening and closing angle of the movable door from the outside to the inside is locked within the time period to prevent other pets from entering; The preset duration is determined based on the historical usage duration of a single pet or a pre-set duration.

Images