A pet health monitoring device and monitoring and early warning system

Abstract

The application discloses a kind of pet health monitoring equipment and monitoring early warning system, it is related to pet health monitoring technical field, including collar main body, sensor group is arranged on the collar main body, camera is arranged on the side of the sensor group, health monitoring chip is built-in in the sensor group, health monitoring system is built-in in the health monitoring chip;The application can dynamically adjust the health threshold range by including environmental parameters and motion state into comprehensive analysis model, fully consider the complex scene of respiratory abnormality of short-nosed dog species under high temperature environment or heart rate change after exercise, thereby substantially reduce the false alarm probability caused by single threshold determination, cooperate with multistage early warning mechanism, such as APP prompt is pushed to one-level slight abnormality, second-grade moderate abnormality sends short message early warning, three-level severe abnormality automatically contacts veterinarian and sends positioning, gradient response of abnormal condition is realized, both ensure the timely processing of emergency condition, also avoid excessive interference user.

Description

Technical Field

[0001] This invention relates to the field of pet health monitoring technology, and in particular to a pet health monitoring device and monitoring and early warning system. Background Technology

[0002] With rising economic levels and changing lifestyles, pets have become important members of many families. Statistics show that pet ownership is increasing globally year by year. Pets not only bring joy to life but also improve the quality of life for family members to some extent. However, this has also led to a growing demand for pet health management. Traditional systems, however, lack dynamic threshold databases linked to pet breed and age, resulting in the use of uniform health standards for judgment and ignoring significant differences in physiological indicators among different breeds and age groups. Furthermore, traditional technologies fail to incorporate environmental parameters and activity levels into their analysis models. For example, they do not consider the fundamental difference between abnormal increases in respiratory rate in short-nosed dogs under high temperatures and increases in heart rate after normal exercise. This single-threshold approach is highly prone to false alarms.

[0003] To address the aforementioned technical deficiencies, a solution is proposed. Summary of the Invention

[0004] The purpose of this invention is to address the problem that traditional systems lack a dynamic threshold database associated with pet breed and age, leading to the use of a uniform health standard for judgment and ignoring significant differences in physiological indicators among pets of different breeds and age groups. Furthermore, traditional technologies fail to incorporate environmental parameters and activity levels into the analysis model, making this single-threshold judgment method highly susceptible to false alarms.

[0005] To achieve the above objectives, the present invention adopts the following technical solution: a pet health monitoring device and monitoring and early warning system, including a collar body, a sensor group is provided on the collar body, a camera is provided on one side of the sensor group, a health monitoring chip is built into the sensor group, and a health monitoring system is built into the health monitoring chip; The health monitoring system includes a data acquisition unit, an attribute recognition unit, an environmental detection unit, a motion detection unit, and a health detection unit; The data acquisition unit is used to detect data in real time during the pet health monitoring process through the sensor group, and integrate the data after denoising and standardization to obtain a dataset. The dataset includes image data, temperature data, light intensity data, humidity data, exercise duration data, exercise distance data, number of jumps and height data, body temperature data, heart rate data, and location coordinate data. The attribute recognition unit determines the pet's species and age group based on images of the pet captured by the camera. The pet's age group is divided into juvenile, growth, and old age. The environmental monitoring unit is used to acquire temperature data, light intensity data, and humidity data, and calculates and analyzes them to obtain the pet's living environment index; The motion detection unit is used to acquire data on exercise duration, distance traveled, number of jumps, and height, and calculates and analyzes these data to determine the pet's exercise index. The health monitoring unit is used to centrally acquire historical health abnormality parameters of pets from the self-data set. These parameters include body temperature data and heart rate data when the pet's health is abnormal. The unit analyzes and obtains a preset fixed health threshold range for the pet. It then combines the living environment index and exercise index to dynamically adjust the preset fixed health threshold range to obtain a dynamic health threshold range. Finally, it calculates a health index based on real-time body temperature data and heart rate data to determine whether the pet is healthy.

[0006] Furthermore, the sensor group is provided with several heat dissipation holes, which are connected to the internal cavity of the sensor group for ventilation and heat dissipation of the health monitoring chip.

[0007] Furthermore, the process for determining the pet's breed and age range is as follows: S21. Collect historical morphological images of different types of pets at different age groups as a dataset, and randomly divide the dataset into a training set and a test set. Label the corresponding age group and pet type on each morphological image in the training set as labels. S22. Construct a pet attribute recognition model based on a convolutional neural network, train the pet attribute recognition model using a training set, and test the pet attribute recognition model using a test set to obtain a qualified pet attribute recognition model. S23. Take a picture of the pet to be monitored using a camera, and input the picture into a qualified pet attribute recognition model to determine the species and age group of the pet to be monitored.

[0008] Furthermore, the analysis and calculation process of the living environment index is as follows: S31. Acquire real-time temperature data, light intensity data, and humidity data, and perform analysis and calculation; S32, through formula Calculate the living environment index ,in, Temperature data for the pet's living environment. Data on light intensity in the pet's living environment. Humidity data for the pet's living environment. The value is a preset standard ambient temperature, which is averaged by summing the upper and lower limits of the suitable temperature range for the corresponding pet's age group and breed. The value is a preset standard ambient light intensity value, which is then averaged by summing the upper and lower limits of the suitable light intensity range for the corresponding pet's age group and breed. The value is a preset standard ambient humidity level, which is then averaged by summing the upper and lower limits of the suitable humidity range for the corresponding pet's age group and breed. The weighting coefficient for ambient temperature. The weighting coefficient for the light intensity of the living environment. The weighting coefficient for ambient humidity is determined by analyzing the impact of extensive temperature, light intensity, and humidity data on pet health. The living environment index is used to indicate the impact of environmental factors on pet health. S33, through formula The lower limit threshold for calculating the living environment index ; through formula Upper threshold for calculating the living environment index ,in, The minimum temperature data of the pet's living environment from the historical pet health dataset. This data represents the minimum light intensity in the living environment of pets from a historical dataset of pets at the time of their health. The minimum humidity data for the pet's living environment from the historical pet health dataset. This is the temperature data of the largest pet's living environment in the historical pet health dataset. This data represents the maximum light intensity in the living environment of pets during historical periods of pet health. The humidity data of the largest pet living environment in the historical pet health dataset is used to form the threshold range of the living environment index. If the living environment index exceeds the threshold range, it indicates that the pet's living environment is abnormal. The abnormal information will be sent to the backend server and a level one minor abnormality warning will be triggered. At the same time, a health reminder will be pushed to the user's terminal through the APP.

[0009] Furthermore, the analysis and calculation process of the exercise index is as follows: S41. Acquire real-time exercise duration data, exercise distance data, and jump count and height data, and perform analysis and calculation; S42, through formula Calculate the fitness index ,in, For pet movement distance data, For pet exercise duration data, This is the pet's jumping data, calculated by multiplying the number of jumps by the jump height. This is based on the maximum range of movement data preset according to the corresponding pet's age group and breed. This is based on the maximum exercise time data preset according to the corresponding pet's age group and breed. This is based on the maximum jump data preset according to the corresponding pet age group and breed. Weighting of pet movement distance Weighting of pet exercise time The weighting of pet jumping is determined by analyzing the impact of extensive data on exercise duration, distance traveled, and the number and height of jumps on pet health. The activity index is used to indicate the impact of pet activity on pet health. The higher the activity index, the more likely the pet is to develop health abnormalities. S43, through formula Calculate the lower threshold of the exercise index ; through formula Upper threshold for calculating the fitness index ,in, This is the minimum pet movement distance data from the historical pet health dataset. The minimum pet exercise duration data in the historical pet health dataset. The minimum pet jump data in the historical pet health dataset. This is the maximum pet movement distance data in the historical pet health dataset. This is the longest pet exercise time recorded in the historical pet health dataset. The maximum pet jump data from the historical pet health dataset is used to define the threshold range for the activity index. If the exercise index exceeds the preset exercise threshold range, it indicates that the pet's exercise is abnormal. The abnormal information will be sent to the backend server and a level two moderate abnormality warning will be triggered. At the same time, a health warning will be sent to the user's terminal via SMS.

[0010] Furthermore, the analysis and calculation process of the health index is as follows: S51. Determine the fixed body temperature threshold range based on the corresponding pet species. and fixed heart rate threshold range ; S52, Combining the living environment index and the activity index, through the formula Calculate the lower limit of dynamic body temperature threshold separately and the upper limit of dynamic body temperature threshold Through formula Calculate the lower limit of dynamic heart rate threshold respectively and dynamic heart rate threshold upper limit ; S53. Acquire real-time body temperature data, heart rate data, as well as living environment index and exercise index, and perform analysis and calculation; S54, through formula Calculate the health index ,in, Provide real-time body temperature data for pets. For real-time heart rate data of pets, This is a correction factor for the pet's living environment on the pet's health index. This is a correction factor for pet exercise on the pet health index, which is used to indicate whether there are any abnormalities in the pet's physical health. S55, through formula The lower threshold for calculating the health index ; through formula Upper limit threshold for calculating health index ,in, The minimum pet body temperature data from the historical pet health dataset. The minimum pet heart rate data from the historical pet health dataset. This is the maximum pet heart rate data from the historical pet health dataset. This is the maximum pet heart rate data from the historical pet health dataset. This dataset contains the average body temperature data of pets from historical pet health records. The average heart rate data of pets in the historical pet health dataset is used to form the threshold range of the living environment index. If the health index exceeds the dynamic health threshold range, it indicates that the pet's health is abnormal. The abnormal information will be sent to the backend server and a level three severe abnormality warning will be triggered. At the same time, a text message will be sent to the user terminal and a preset veterinarian's phone number will be automatically dialed, and the location coordinate data will be sent to issue a health alert.

[0011] In summary, due to the adoption of the above technical solution, the beneficial effects of the present invention are: This pet health monitoring device and early warning system incorporates environmental parameters (such as temperature, light, and humidity) and activity status (such as exercise duration, distance, and jump data) into a comprehensive analysis model. This allows for dynamic adjustment of health threshold ranges, fully considering complex scenarios such as respiratory abnormalities in short-nosed breeds under high-temperature conditions or changes in heart rate after exercise. This significantly reduces the probability of false alarms caused by single threshold judgments. Furthermore, using convolutional neural networks for pet image attribute recognition not only quickly and accurately determines breed and age group but also continuously learns and optimizes the model, further enhancing the system's adaptability and intelligence. Coupled with a multi-level early warning mechanism (e.g., Level 1 push notifications to the app for minor abnormalities, Level 2 SMS alerts for moderate abnormalities, and Level 3 automatic contact with a veterinarian and location updates for severe abnormalities), a tiered response to abnormal situations is achieved. This ensures timely handling of emergencies while avoiding excessive user interference, significantly improving the user experience. Moreover, by introducing comparative analysis of historical health data and real-time monitoring results, the system can dynamically adjust threshold ranges and generate a health index. Combining the comprehensive influence of environmental and activity factors, it provides a multi-dimensional scientific assessment of pet health status, effectively improving the efficiency of pet health monitoring. Attached Figure Description

[0012] Figure 1 A schematic diagram of the system flow of the present invention is shown; Figure 2 A schematic diagram of the pet collar of the present invention is shown; Figure 3 A schematic diagram of the sensor group of the present invention is shown; Legend: 1. Necklace body; 2. Sensor group; 3. Camera; 4. Heat dissipation holes; 5. Health monitoring chip. Detailed Implementation

[0013] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0014] Example 1: like Figure 1 As shown, a pet health monitoring and early warning system includes a data acquisition unit, an attribute recognition unit, an environmental detection unit, an exercise detection unit, and a health detection unit. The data acquisition unit is used to detect data in real time during the pet health monitoring process through sensor group 2, and integrate the data after denoising and standardization to obtain a dataset. The dataset includes image data, temperature data, light intensity data, humidity data, exercise duration data, exercise distance data, number of jumps and height data, body temperature data, heart rate data, and positioning coordinate data. The attribute recognition unit determines the pet's species and age group based on the pet's image captured by camera 3. The pet's age group is divided into juvenile, growth, and old age. The process for determining the breed and age range of a pet is as follows: S21. Collect historical morphological images of different types of pets at different age groups as a dataset, and randomly divide the dataset into a training set and a test set. Label the corresponding age group and pet type on each morphological image in the training set as labels. S22. Construct a pet attribute recognition model based on a convolutional neural network, train the pet attribute recognition model using a training set, and test the pet attribute recognition model using a test set to obtain a qualified pet attribute recognition model. S23. Take a morphological image of the pet to be monitored using camera 3, and input the morphological image into a qualified pet attribute recognition model to determine the species and age group of the pet to be monitored.

[0015] The environmental monitoring unit is used to acquire temperature data, light intensity data, and humidity data, and calculates and analyzes them to obtain the pet's living environment index; The analysis and calculation process of the living environment index is as follows: S31. Acquire real-time temperature data, light intensity data, and humidity data, and perform analysis and calculation; S32, through formula Calculate the living environment index ,in, Temperature data for the pet's living environment. Data on light intensity in the pet's living environment. Humidity data for the pet's living environment. The value is a preset standard ambient temperature, which is averaged by summing the upper and lower limits of the suitable temperature range for the corresponding pet's age group and breed. The value is a preset standard ambient light intensity value, which is then averaged by summing the upper and lower limits of the suitable light intensity range for the corresponding pet's age group and breed. The value is a preset standard ambient humidity level, which is then averaged by summing the upper and lower limits of the suitable humidity range for the corresponding pet's age group and breed. The weighting coefficient for ambient temperature. The weighting coefficient for the light intensity of the living environment. The weighting coefficient for ambient humidity is determined by analyzing the impact of extensive temperature, light intensity, and humidity data on pet health. The living environment index is used to indicate the impact of environmental factors on pet health. S33, through formula The lower limit threshold for calculating the living environment index ; through formula Upper threshold for calculating the living environment index ,in, The minimum temperature data of the pet's living environment from the historical pet health dataset. This data represents the minimum light intensity in the living environment of pets from a historical dataset of pets at the time of their health. The minimum humidity data for the pet's living environment from the historical pet health dataset. This is the temperature data of the largest pet's living environment in the historical pet health dataset. This data represents the maximum light intensity in the living environment of pets during historical periods of pet health. The humidity data of the largest pet living environment in the historical pet health dataset is used to form the threshold range of the living environment index. If the living environment index exceeds the threshold range, it indicates that the pet's living environment is abnormal. The abnormal information will be sent to the backend server and a first-level minor abnormality warning will be triggered. At the same time, a health reminder will be pushed to the user terminal through the APP. The backend server (such as a remote computer platform) is used to remotely and wirelessly receive data information from the health monitoring chip 5. At the same time, it can send information to the sending user terminal (such as a mobile phone or tablet).

[0016] The motion detection unit is used to acquire data on exercise duration, distance traveled, number of jumps, and height, and calculates and analyzes these data to determine the pet's exercise index. The analysis and calculation process of the exercise index is as follows: S41. Acquire real-time exercise duration data, exercise distance data, and jump count and height data, and perform analysis and calculation; S42, through formula Calculate the fitness index ,in, For pet movement distance data, For pet exercise duration data, This is the pet's jumping data, calculated by multiplying the number of jumps by the jump height. This is based on the maximum range of movement data preset according to the corresponding pet's age group and breed. This is based on the maximum exercise time data preset according to the corresponding pet's age group and breed. This is based on the maximum jump data preset according to the corresponding pet age group and breed. Weighting of pet movement distance Weighting of pet exercise time The weighting of pet jumping is determined by analyzing the impact of extensive data on exercise duration, distance traveled, and the number and height of jumps on pet health. The activity index is used to indicate the impact of pet activity on pet health. The higher the activity index, the more likely the pet is to develop health abnormalities. S43, through formula Calculate the lower threshold of the exercise index ; through formula Upper threshold for calculating the fitness index ,in, This is the minimum pet movement distance data from the historical pet health dataset. The minimum pet exercise duration data in the historical pet health dataset. The minimum pet jump data in the historical pet health dataset. This is the maximum pet movement distance data in the historical pet health dataset. This is the longest pet exercise time recorded in the historical pet health dataset. The maximum pet jump data from the historical pet health dataset is used to define the threshold range for the activity index. If the exercise index exceeds the preset exercise threshold range, it indicates that the pet's exercise is abnormal. The abnormal information will be sent to the backend server and a level two moderate abnormality warning will be triggered. At the same time, a health warning will be sent to the user's terminal via SMS.

[0017] The health monitoring unit is used to centrally acquire historical health abnormality parameters of pets from the self-data set. The historical health abnormality parameters of pets include body temperature data and heart rate data when the pet's health is abnormal. The unit analyzes and obtains a preset fixed health threshold range for pets. It then combines the living environment index and exercise index to dynamically adjust the preset fixed health threshold range to obtain a dynamic health threshold range. Finally, it calculates a health index based on real-time body temperature data and heart rate data to determine whether the pet is healthy. The analysis and calculation process of the health index is as follows: S51. Determine the fixed body temperature threshold range based on the corresponding pet species. and fixed heart rate threshold range ; S52, Combining the living environment index and the activity index, through the formula Calculate the lower limit of dynamic body temperature threshold separately and the upper limit of dynamic body temperature threshold Through formula Calculate the lower limit of dynamic heart rate threshold respectively and dynamic heart rate threshold upper limit ; S53. Acquire real-time body temperature data, heart rate data, as well as living environment index and exercise index, and perform analysis and calculation; S54, through formula Calculate the health index ,in, Provide real-time body temperature data for pets. For real-time heart rate data of pets, This is a correction factor for the pet's living environment on the pet's health index. This is a correction factor for pet exercise on the pet health index, which is used to indicate whether there are any abnormalities in the pet's physical health. S55, through formula The lower threshold for calculating the health index ; through formula Upper limit threshold for calculating health index ,in, The minimum pet body temperature data from the historical pet health dataset. The minimum pet heart rate data from the historical pet health dataset. This is the maximum pet heart rate data from the historical pet health dataset. This is the maximum pet heart rate data from the historical pet health dataset. This dataset contains the average body temperature data of pets from historical pet health records. The average heart rate data of pets in the historical pet health dataset is used to form the threshold range of the living environment index. If the health index exceeds the dynamic health threshold range, it indicates that the pet's health is abnormal. The abnormal information will be sent to the backend server and a level three severe abnormality warning will be triggered. At the same time, a text message will be sent to the user terminal and a preset veterinarian's phone number will be automatically dialed, and the location coordinate data will be sent to issue a health alert.

[0018] By incorporating environmental parameters (such as temperature, light, and humidity) and activity status (such as exercise duration, distance, and jump data) into a comprehensive analysis model, the system can dynamically adjust the health threshold range. This fully considers complex scenarios such as respiratory abnormalities in short-nosed dogs under high-temperature conditions or changes in heart rate after exercise, significantly reducing the probability of false alarms caused by single-threshold judgments. Furthermore, using convolutional neural networks for attribute recognition of pet images not only quickly and accurately determines breed and age group but also continuously learns and optimizes the model, further enhancing the system's adaptability and intelligence. Coupled with a multi-level early warning mechanism (such as a push notification to the app for minor anomalies, an SMS alert for moderate anomalies, and automatic contact with a veterinarian and location updates for severe anomalies), a tiered response to abnormal situations is achieved. This ensures timely handling of emergencies while avoiding excessive user interference, significantly improving the user experience. Moreover, by introducing comparative analysis of historical health data and real-time monitoring results, the system can dynamically adjust the threshold range and generate a health index. Combining the comprehensive influence of environmental and activity factors, it provides a multi-dimensional scientific assessment of pet health status, effectively improving the efficiency of pet health monitoring.

[0019] Example 2: like Figure 2-3 As shown, a pet health monitoring device includes a collar body 1, a sensor group 2 on the collar body 1, a camera 3 on one side of the sensor group 2, a built-in health monitoring chip 5 in the sensor group 2, and a built-in health monitoring system in the health monitoring chip 5 for monitoring the pet's health. When the pet's health is abnormal, it will issue an early warning to the user. Several heat dissipation holes 4 are provided on the sensor group 2, which are connected to the internal cavity of the sensor group 2 for ventilation and heat dissipation of the health monitoring chip 5.

[0020] The size of the interval and threshold is set to facilitate comparison. The size of the threshold depends on the amount of sample data and the number of bases set by those skilled in the art for each set of sample data; as long as it does not affect the ratio between the parameter and the quantized value.

[0021] The above formulas are all dimensionless calculations. The formulas are derived from software simulations based on a large amount of collected data to obtain the most recent real-world results. The preset parameters in the formulas are set by those skilled in the art according to the actual situation. In the two embodiments provided in this application, it should be understood that the disclosed apparatus and system can be implemented in other ways; for example, the apparatus embodiments described above are merely illustrative, for example, the division of modules is merely a logical functional division, and there may be other division methods in actual implementation, such as multiple modules or components can be combined or integrated into another system, or some features can be ignored or not executed; another point is that the mutual coupling or direct coupling or communication connection shown or discussed can be through some interfaces, and the indirect coupling or communication connection of the apparatus or module can be electrical, mechanical or other forms. The above description is only a preferred embodiment of the present invention, but the scope of protection of the present invention is not limited thereto. Any equivalent substitutions or modifications made by those skilled in the art within the scope of the technology disclosed in the present invention, based on the technical solution and inventive concept of the present invention, should be covered within the scope of protection of the present invention.

Claims

1. A pet health monitoring device and monitoring and early warning system, characterized in that, The collar body (1) is characterized in that a sensor group (2) is provided on the collar body (1), a camera (3) is provided on one side of the sensor group (2), the sensor group (2) has a built-in health monitoring chip (5), and the health monitoring chip (5) has a built-in health monitoring system. The health monitoring system includes a data acquisition unit, an attribute recognition unit, an environmental detection unit, a motion detection unit, and a health detection unit; The data acquisition unit is used to detect data in real time during the pet health monitoring process through the sensor group (2), and integrate the data after denoising and standardization to obtain a dataset. The dataset includes image data, temperature data, light intensity data, humidity data, exercise duration data, exercise distance data, number of jumps and height data, body temperature data, heart rate data and positioning coordinate data. The attribute recognition unit determines the pet's species and age group based on the pet's image captured by the camera (3), wherein the pet's age group is divided into juvenile, growth and old age. The environmental monitoring unit is used to acquire temperature data, light intensity data, and humidity data, and calculates and analyzes them to obtain the pet's living environment index; The motion detection unit is used to acquire data on exercise duration, distance traveled, number of jumps, and height, and calculates and analyzes these data to determine the pet's exercise index. The health monitoring unit is used to centrally acquire historical health abnormality parameters of pets from the self-data set. The historical health abnormality parameters of pets include body temperature data and heart rate data when the pet's health is abnormal. The unit analyzes and obtains a preset fixed health threshold range for pets. It then combines the living environment index and exercise index to dynamically adjust the preset fixed health threshold range to obtain a dynamic health threshold range. Finally, it calculates the health index based on real-time body temperature data and heart rate data to determine whether the pet is healthy. The analysis and calculation process of the health index is as follows: S51. Determine the fixed body temperature threshold range based on the corresponding pet species. and fixed heart rate threshold range ; S52, Combining the living environment index and the activity index, through the formula Calculate the lower limit of dynamic body temperature threshold separately and the upper limit of dynamic body temperature threshold Through formula Calculate the lower limit of dynamic heart rate threshold respectively and dynamic heart rate threshold upper limit ; S53. Acquire real-time body temperature data, heart rate data, as well as living environment index and exercise index, and perform analysis and calculation; S54, through formula Calculate the health index ,in, Provide real-time body temperature data for pets. For real-time heart rate data of pets, This is a correction factor for the pet's living environment on the pet's health index. This is a correction factor for pet exercise on the pet health index, which is used to indicate whether there are any abnormalities in the pet's physical health. S55, through formula The lower threshold for calculating the health index ; through formula Upper limit threshold for calculating health index ,in, The minimum pet body temperature data from the historical pet health dataset. The minimum pet heart rate data from the historical pet health dataset. This is the maximum pet heart rate data from the historical pet health dataset. This is the maximum pet heart rate data from the historical pet health dataset. This dataset contains the average body temperature data of pets from historical pet health records. The average heart rate data of pets in the historical pet health dataset is used to form the threshold range of the living environment index. If the health index exceeds the dynamic health threshold range, it indicates that the pet's health is abnormal. The abnormal information will be sent to the backend server and a level three severe abnormality warning will be triggered. At the same time, a text message will be sent to the user terminal and a preset veterinarian's phone number will be automatically dialed, and the location coordinate data will be sent to issue a health alert.

2. The pet health monitoring device and monitoring and early warning system according to claim 1, characterized in that, The sensor group (2) is provided with several heat dissipation holes (4), which are connected to the internal cavity of the sensor group (2) for ventilation and heat dissipation of the health monitoring chip (5).

3. The pet health monitoring device and monitoring and early warning system according to claim 1, characterized in that, The process for determining the breed and age range of a pet is as follows: S21. Collect historical morphological images of different types of pets at different age groups as a dataset, and randomly divide the dataset into a training set and a test set. Label the corresponding age group and pet type on each morphological image in the training set as labels. S22. Construct a pet attribute recognition model based on a convolutional neural network, train the pet attribute recognition model using a training set, and test the pet attribute recognition model using a test set to obtain a qualified pet attribute recognition model. S23. Take a morphological image of the pet to be monitored through the camera (3) and input the morphological image into the pet attribute recognition model that has passed the test to determine the type and age range of the pet to be monitored.

4. The pet health monitoring device and monitoring and early warning system according to claim 1, characterized in that, The analysis and calculation process of the living environment index is as follows: S31. Acquire real-time temperature data, light intensity data, and humidity data, and perform analysis and calculation; S32, through formula Calculate the living environment index ,in, Temperature data for the pet's living environment. Data on light intensity in the pet's living environment. Humidity data for the pet's living environment. The value is a preset standard ambient temperature, which is averaged by summing the upper and lower limits of the suitable temperature range for the corresponding pet's age group and breed. The value is a preset standard ambient light intensity value, which is then averaged by summing the upper and lower limits of the suitable light intensity range for the corresponding pet's age group and breed. The value is a preset standard ambient humidity level, which is then averaged by summing the upper and lower limits of the suitable humidity range for the corresponding pet's age group and breed. The weighting coefficient for ambient temperature. The weighting coefficient for the light intensity of the living environment. The weighting coefficient for ambient humidity is determined by analyzing the impact of extensive temperature, light intensity, and humidity data on pet health. The living environment index is used to indicate the impact of environmental factors on pet health. S33, through formula The lower limit threshold for calculating the living environment index ; through formula Upper threshold for calculating the living environment index ,in, The minimum temperature data of the pet's living environment from the historical pet health dataset. This data represents the minimum light intensity in the living environment of pets from a historical dataset of pets at the time of their health. The minimum humidity data for the pet's living environment from the historical pet health dataset. This is the temperature data of the largest pet's living environment in the historical pet health dataset. This data represents the maximum light intensity in the living environment of pets during historical periods of pet health. The humidity data of the largest pet living environment in the historical pet health dataset is used to form the threshold range of the living environment index. If the living environment index exceeds the threshold range, it indicates that the pet's living environment is abnormal. The abnormal information will be sent to the backend server and a level one minor abnormality warning will be triggered. At the same time, a health reminder will be pushed to the user's terminal through the APP.

5. The pet health monitoring device and monitoring and early warning system according to claim 1, characterized in that, The analysis and calculation process of the exercise index is as follows: S41. Acquire real-time exercise duration data, exercise distance data, and jump count and height data, and perform analysis and calculation; S42, through formula Calculate the fitness index ,in, For pet movement distance data, For pet exercise duration data, This is the pet's jumping data, calculated by multiplying the number of jumps by the jump height. This is based on the maximum range of movement data preset according to the corresponding pet's age group and breed. This is based on the maximum exercise time data preset according to the corresponding pet's age group and breed. This is based on the maximum jump data preset according to the corresponding pet age group and breed. Weighting of pet movement distance Weighting of pet exercise time The weighting of pet jumping is determined by analyzing the impact of extensive data on exercise duration, distance traveled, and the number and height of jumps on pet health. The activity index is used to indicate the impact of pet activity on pet health. The higher the activity index, the more likely the pet is to develop health abnormalities. S43, through formula Calculate the lower threshold of the exercise index ; through formula Upper threshold for calculating the fitness index ,in, This is the minimum pet movement distance data from the historical pet health dataset. The minimum pet exercise duration data in the historical pet health dataset. The minimum pet jump data in the historical pet health dataset. This is the maximum pet movement distance data in the historical pet health dataset. This is the longest pet exercise time recorded in the historical pet health dataset. The maximum pet jump data from the historical pet health dataset is used to define the threshold range for the activity index. If the exercise index exceeds the preset exercise threshold range, it indicates that the pet's exercise is abnormal. The abnormal information will be sent to the backend server and a level two moderate abnormality warning will be triggered. At the same time, a health warning will be sent to the user's terminal via SMS.

Images