Abnormal body temperature detection method and related equipment thereof

Abstract

The invention relates to an abnormal body temperature detection method and related equipment thereof. The method comprises the steps that the body surface temperature of a measured object is measured, and corresponding time information is recorded; the environment temperature of the environment where the measured object is located is measured, and corresponding time information is recorded; the future predicted body surface temperature of the measured object is predicted based on the measured body surface temperature observed in the past of the measured object and known environment temperature and time information, and the time information comprises quarter, month and intra-day moments; on the basis of the predicted body surface temperature and the measured body surface temperature observed in the future, body temperature abnormity of the measured object is detected. Through the method and the device, the problem that time and labor are wasted in abnormal body temperature detection of livestock farm animals in the prior art is solved, the labor cost is saved, and the abnormal body temperature early warning effect of the livestock farm animals is improved.

Description

Technical Field

[0001] The present application relates to the field of animal husbandry technology, and in particular to an abnormal body temperature detection method and related equipment. Background Art

[0002] This section is intended to provide a background or context to the embodiments of the invention recited in the claims. No admission is made that the description herein is prior art by its inclusion in this section.

[0003] The most feared thing in the agricultural breeding industry is that animals get sick, and fever is a common manifestation of many diseases. Automatic abnormal body temperature detection of farmed animals has become an urgent need. However, in related technologies, farmers still measure the rectal temperature of animals manually, which is time-consuming and laborious. At the same time, due to the lack of large-scale rectal temperature data, common prediction technologies cannot be used. Summary of the invention

[0004] The abnormal body temperature detection method and related equipment provided by the embodiments of the present invention at least solve the problem that abnormal body temperature detection of farm animals is time-consuming and labor-intensive in the related art.

[0005] An abnormal body temperature detection system, comprising: A body surface temperature measuring device, used to measure the body surface temperature of the object being measured; An ambient temperature measuring device, used to measure the ambient temperature of the environment where the measured object is located; A body surface temperature prediction device, used to predict the predicted body surface temperature of the measured object in the future based on a first time-varying variable observed in the past and a known second time-varying variable of the measured object, wherein the first time-varying variable includes the measured body surface temperature observed in the past, and the second time-varying variable includes the ambient temperature and time information, and the time information includes: quarter, month and time of day; The abnormal body temperature detection device is used to detect the abnormal body temperature of the object under test based on the predicted body surface temperature and the measured body surface temperature observed in the future.

[0006] In some embodiments, the body surface temperature prediction device comprises: A static variable encoder, used for encoding a static variable into a static feature vector, wherein the static variable includes attribute information of the measured object; A time-varying variable encoder, used for encoding the first time-varying variable and the second time-varying variable into a first time-varying feature vector; A time series converter is used to process the static feature vector and the first time-varying feature vector based on pre-trained parameters to obtain a second time-varying feature vector for representing the predicted future body surface temperature of the measured object.

[0007] In some of the embodiments, the temporal converter includes: a gated residual network for adjusting the weights of the static feature vector and the first time-varying feature vector during feature fusion.

[0008] In some embodiments, the abnormal body temperature detection device includes: A mutual correlation coefficient calculation module, used to calculate the mutual correlation coefficient between the predicted body surface temperature and the measured body surface temperature observed in the future; The abnormal body temperature detection module is used to generate an abnormal body temperature detection result of the object under test based on the mutual correlation coefficient.

[0009] In some embodiments, the abnormal body temperature detection device further includes: The fluctuation range amplification module is used to increase the fluctuation range of the predicted body surface temperature after normalization and the measured body surface temperature observed in the future.

[0010] In some of the embodiments, the time span between the predicted body surface temperature and the measured body surface temperature to be observed in the future ranges from one to three hours.

[0011] A body surface temperature prediction device, comprising: A static variable encoder, used for encoding static variables of the measured object into static feature vectors, wherein the static variables include attribute information of the measured object; A time-varying variable encoder, used for encoding a first time-varying variable observed in the past and a known second time-varying variable of the measured object into a first time-varying feature vector, wherein the first time-varying variable includes a measured body surface temperature observed in the past, and the second time-varying variable includes an ambient temperature and time information, and the time information includes: quarter, month and time of day; A time series converter is used to process the static feature vector and the first time-varying feature vector based on pre-trained parameters to obtain a second time-varying feature vector for representing the predicted future body surface temperature of the measured object.

[0012] In some of the embodiments, the temporal converter includes: a gated residual network for adjusting the weights of the static feature vector and the first time-varying feature vector during feature fusion.

[0013] A method for detecting abnormal body temperature, comprising: Measure the body surface temperature of the object being measured and record the corresponding time information; Measuring the ambient temperature of the environment where the measured object is located and recording the corresponding time information; Based on the measured body surface temperature of the measured object observed in the past, as well as the known ambient temperature and time information, predict the predicted body surface temperature of the measured object in the future, wherein the time information includes: quarter, month and time of day; Based on the predicted body surface temperature and the measured body surface temperature observed in the future, abnormal body temperature of the object being measured is detected.

[0014] In some embodiments, based on the predicted body surface temperature and the measured body surface temperature observed in the future, detecting abnormal body temperature of the subject includes: Normalizing the predicted body surface temperature and the measured body surface temperature observed in the future, and increasing the fluctuation range of the processed body surface temperature; Calculating a correlation coefficient between the predicted body surface temperature and the measured body surface temperature observed in the future; Based on the mutual correlation coefficient, a body temperature abnormality detection result of the measured object is generated.

[0015] The abnormal body temperature detection method and related equipment provided by the embodiments of the present invention measure the body surface temperature of the object to be measured and record the corresponding time information; measure the ambient temperature of the environment where the object to be measured is located and record the corresponding time information; predict the predicted future body surface temperature of the object to be measured based on the measured body surface temperature observed in the past, and the known ambient temperature and time information, wherein the time information includes: quarter, month and time of day; based on the predicted body surface temperature and the measured body surface temperature observed in the future, the abnormal body temperature of the object to be measured is detected, which solves the problem of time-consuming and labor-intensive abnormal body temperature detection of farm animals in related technologies, saves labor costs, and improves the abnormal body temperature warning effect of farm animals.

[0016] The details of one or more embodiments of the invention are set forth in the following drawings and description so that other features, objects, and advantages of the invention are more readily apparent. BRIEF DESCRIPTION OF THE DRAWINGS

[0017] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the following briefly introduces the drawings required for use in the embodiments or the prior art descriptions. Obviously, the drawings described below are only some embodiments of the present invention, and for ordinary technicians in this field, other embodiments can be obtained based on these drawings without creative work.

[0018] Figure 1 4 is a flow chart of the abnormal body temperature detection method of this embodiment.

[0019] Figure 2 Schematic diagram of body surface temperature and ambient temperature collected by the same ear tag of this embodiment at different times of the year.

[0020] Figure 3 Schematic diagram of the structure of the abnormal body temperature detection system of this embodiment.

[0021] Figure 4 Schematic diagram of the structure of the body surface temperature prediction device of this embodiment.

[0022] Figure 5 Schematic diagram of the structure of the electronic device of this embodiment. DETAILED DESCRIPTION

[0023] Embodiments of the present embodiment will be described in more detail below with reference to the accompanying drawings. Although certain embodiments of the present embodiment are shown in the accompanying drawings, it should be understood that the present embodiment can be implemented in various forms and should not be construed as being limited to the embodiments set forth herein, which are instead provided for a more thorough and complete understanding of the present embodiment. It should be understood that the drawings and embodiments of the present embodiment are only for exemplary purposes and are not intended to limit the scope of protection of the present embodiment.

[0024] Currently in the breeding industry, the rectal temperature measured at the anus of animals is a reliable indicator for measuring animal body temperature. Abnormal animal body temperature is manifested externally as abnormal measured rectal temperature. However, in the related technology, since there is still no solution to set a temperature measurement device inside the animal or on the anus, the rectal temperature of animals mainly relies on manual measurement. On the one hand, manual measurement of rectal temperature requires a lot of manpower. On the other hand, the manually measured rectal temperature is only one or several measurement values ​​at the time of measurement. A single or several measurement values ​​are greatly affected by errors or other accidental factors, and abnormal body temperature is prone to omission or false alarm.

[0025] Based on the deep learning model, using one or several rectal temperatures at the measurement time to predict whether the animal's body temperature is abnormal is also a possible abnormal body temperature detection scheme. However, this scheme relies on continuous and long-term detection of the animal's rectal temperature. Currently, there is no technology that can achieve the goal of continuous and long-term detection of the animal's rectal temperature, so there is a lack of sample data for training the above deep learning model.

[0026] Compared with the difficulty of continuously measuring the rectal temperature of animals, the surface temperature of animals is easier to continuously collect. For example, ear tag sensors are currently used to collect the temperature inside the ears of animals around the clock. The temperature inside the ear is referred to as ear temperature. Ear temperature reflects the surface temperature of animals and is more easily affected by the ambient temperature than rectal temperature. Therefore, it is impossible to directly use the ear temperature of animals to detect whether the animal's body temperature is abnormal.

[0027] In order to solve the above problems in the related art, an abnormal body temperature detection method is provided in this embodiment. Figure 1is a flow chart of the abnormal body temperature detection method of this embodiment. Figure 1 As shown, the process includes the following steps: Step S101, measuring the body surface temperature of the object being measured and recording the corresponding time information.

[0028] The surface temperature of the object to be measured can be measured based on a temperature measuring device attached to the surface of the animal. For example, the temperature measuring device can be an ear tag, an animal foot ring or other device with a temperature measurement function. The temperature measuring device can have a time device and a function of recording time. The temperature measuring device can record the time of the measured temperature while measuring the temperature, and transmit the temperature time data to the server through wireless communication (including but not limited to cellular network or wireless local area network, etc.). The temperature measuring device may also not have a time device and the function of recording time. The temperature measuring device transmits the measured temperature data to the server in real time through wireless communication. The server determines the time information corresponding to the temperature data based on the time when the temperature data is received.

[0029] Step S102, measuring the ambient temperature of the environment where the measured object is located and recording the corresponding time information.

[0030] Among them, the environment where the measured object is located, for example, when the measured object is a cow, the environment where it is located can be an open or closed cowshed, pasture, etc. The ambient temperature of the environment where the cow is located refers to the temperature of the open or closed cowshed or pasture. Especially for the pasture, its ambient temperature can be directly derived from the meteorological data of the area where the pasture is located, or it can be collected by a temperature sensor set up in the cowshed or pasture. The temperature measuring device for measuring the ambient temperature can have a time device and a function of recording time. The temperature measuring device can record the time of the measured temperature while measuring the temperature, and transmit the temperature time data to the server by wireless communication (including but not limited to cellular network or wireless local area network, etc.). The temperature measuring device may also not have the function of a time device and recording time. The temperature measuring device transmits the measured temperature data to the server in real time by wireless communication. The server determines the time information corresponding to the temperature data based on the time when the temperature data is received.

[0031] Step S103, predicting the future predicted body surface temperature of the measured object based on the measured body surface temperature observed in the past, the known ambient temperature and time information, wherein the time information includes: quarter, month and time of day.

[0032] In the above steps, the future body surface temperature can be predicted based on a deep learning model trained by measuring the body surface temperature, ambient temperature and time information of animals with normal rectal temperature in the past. The deep learning model can be, for example, a deep learning model based on a temporal-transformer framework.

[0033] The above time information includes quarter, month and time of day. Figure 2 is a schematic diagram of the body surface temperature and the ambient temperature collected by the same ear tag of this embodiment at different times of the year, such as Figure 2 As shown in the figure, the gray curve represents the ambient temperature, and the yellow curve represents the animal's body surface temperature. The body surface temperature measured in spring and summer is relatively stable, while the body surface temperature measured in autumn and winter will fluctuate dramatically with the temperature difference between day and night. This is because the temperature measuring and thermal conductive material of the ear tag is easily affected by changes in ambient temperature, and is affected by ambient temperature to different degrees in different seasons and at different times of the day (i.e., 24 hours a day). Therefore, by also using time information as an input feature of the deep learning model, the model can learn the degree to which the body surface temperature and ambient temperature are affected at different times, thereby achieving accurate prediction of the body surface temperature.

[0034] Step S104, detecting abnormal body temperature of the subject based on the predicted body surface temperature and the measured body surface temperature observed in the future.

[0035] When the predicted body surface temperature is predicted and the measured body surface temperature at the time (or time period) corresponding to the predicted body surface temperature is collected, by comparing the deviation between the two body surface temperatures, it can be known whether the body temperature of the object being measured is abnormal. Among them, the predicted body surface temperature is predicted by the measured body surface temperature of an animal with normal body temperature and the measured ambient temperature. When the model used for the prediction converges, it is considered that the predicted body surface temperature is the normal body surface temperature that should be at the current ambient temperature and the current time. If the measured body surface temperature measured at the current ambient temperature and the current time deviates too much from the normal body surface temperature, it can be considered that the body temperature of the object being measured is abnormal. After the body temperature is abnormal, the worker can be notified to measure the rectal temperature manually to further confirm whether the body temperature of the object being measured is normal.

[0036] Through the above steps, unlike the related art in which the anal temperature is directly collected to detect the abnormal body temperature of the animal, in this embodiment, the correlation between the animal's body surface temperature, the ambient temperature and the time information is learned, and the abnormal body temperature of the animal is detected by detecting the animal's body surface temperature. The animal's body surface temperature, the ambient temperature and the time information can all be obtained conveniently and continuously, thereby reducing the labor cost and avoiding the omission caused by the untimely manual detection of the anal temperature. At the same time, in the above step S104, the surface temperature prediction value sequence and the surface temperature measurement value sequence of the animal's body temperature for a period of time in the future can be compared, thereby achieving more reliable abnormal body temperature detection, avoiding the omission or false alarm of abnormal animal body temperature caused by a single measurement of the animal's anal temperature.

[0037] Among them, in each abnormal body temperature detection, the time span of the predicted body surface temperature and the measured body surface temperature observed in the future ranges from one to three hours. For example, in each prediction step, the predicted body surface temperature of the current step is predicted based on the measured body surface temperature observed in the past at the current moment, the past ambient temperature and time information, and the ambient temperature and time information at the current moment. Therefore, the ambient temperature and time information at the current moment are both known time-varying variables. At the same time, the measured body surface temperature is measured at the current moment. When the number of steps of the predicted predicted body surface temperature reaches the set number of steps, for example, the time corresponding to the accumulated number of steps reaches the above-mentioned time span (for example, three hours), the time series of the predicted body surface temperature and the measured body surface temperature at each step is used as the object of animal body temperature abnormality detection, and an abnormal animal body temperature detection is realized.

[0038] Since the variation range of body surface temperature always varies within a relatively small range, in order to be able to more accurately compare the difference between the time series of predicted body surface temperature and measured body surface temperature within a certain period of time, in some embodiments, based on the predicted body surface temperature and the measured body surface temperature observed in the future, the abnormal body temperature of the object under test is detected, including: normalizing the predicted body surface temperature and the measured body surface temperature observed in the future, and increasing the fluctuation range of the processed body surface temperature; calculating the mutual correlation coefficient between the predicted body surface temperature and the measured body surface temperature observed in the future; based on the mutual correlation coefficient, generating the abnormal body temperature detection result of the object under test. The above embodiment is performed by normalizing the measured body surface temperature and the predicted body surface temperature, then increasing the fluctuation range of the body surface temperature, and finally calculating the mutual correlation coefficient of the two sequences after the fluctuation range increase processing.

[0039] The mutual correlation coefficient can be expressed as: Among them, x i is the ith predicted body surface temperature, y i is the i-th measured body surface temperature.

[0040] The value range of the mutual correlation coefficient is between -1 and +1, and the positive or negative nature represents the correlation between the two sequences (positive / negative correlation). When the animal's body temperature is normal, the mutual correlation coefficient should be close to 1; if the mutual correlation coefficient is close to 0 or is negative, the animal is likely to have a fever and an abnormal body temperature.

[0041] Among them, the fluctuation range of the sequence can be increased by the following methods: Among them, x is the original body temperature value, α and β are constants. For example, α is 0.5 and β is 10.

[0042] This embodiment also provides an abnormal body temperature detection system. Figure 3 : is a schematic diagram of the structure of the abnormal body temperature detection system of this embodiment. Figure 3 As shown, the system includes: The body surface temperature measuring device 30 is used to measure the body surface temperature of the object being measured.

[0043] The surface temperature of the object to be measured can be measured based on a temperature measuring device attached to the surface of the animal. For example, the temperature measuring device can be an ear tag, an animal foot ring or other device with a temperature measurement function. The temperature measuring device can have a time device and a function of recording time. The temperature measuring device can record the time of the measured temperature while measuring the temperature, and transmit the temperature time data to the server through wireless communication (including but not limited to cellular network or wireless local area network, etc.). The temperature measuring device may also not have a time device and the function of recording time. The temperature measuring device transmits the measured temperature data to the server in real time through wireless communication. The server determines the time information corresponding to the temperature data based on the time when the temperature data is received.

[0044] The ambient temperature measuring device 31 is used to measure the ambient temperature of the environment where the measured object is located.

[0045] Among them, the environment where the measured object is located, for example, when the measured object is a cow, the environment where it is located can be an open or closed cowshed, pasture, etc. The ambient temperature of the environment where the cow is located refers to the temperature of the open or closed cowshed or pasture. Especially for the pasture, its ambient temperature can be directly derived from the meteorological data of the area where the pasture is located, or it can be collected by a temperature sensor set up in the cowshed or pasture. The temperature measuring device for measuring the ambient temperature can have a time device and a function of recording time. The temperature measuring device can record the time of the measured temperature while measuring the temperature, and transmit the temperature time data to the server by wireless communication (including but not limited to cellular network or wireless local area network, etc.). The temperature measuring device may also not have the function of a time device and recording time. The temperature measuring device transmits the measured temperature data to the server in real time by wireless communication. The server determines the time information corresponding to the temperature data based on the time when the temperature data is received.

[0046] The body surface temperature prediction device 32 is used to predict the future predicted body surface temperature of the object under test based on the first time-varying variable observed in the past and the known second time-varying variable of the object under test, wherein the first time-varying variable includes the measured body surface temperature observed in the past, and the second time-varying variable includes the ambient temperature and time information, and the time information includes: quarter, month and time of day.

[0047] The body surface temperature prediction device 32 may be a deep learning model trained based on the measured body surface temperature, measured ambient temperature and time information of animals with normal rectal temperature in a historical period. The deep learning model may be, for example, a deep learning model based on a temporal-transformer framework.

[0048] The abnormal body temperature detection device 33 is used to detect abnormal body temperature of the object under test based on the predicted body surface temperature and the measured body surface temperature observed in the future.

[0049] The abnormal body temperature detection device 33 includes: a correlation coefficient calculation module for calculating the correlation coefficient between the predicted body surface temperature and the measured body surface temperature observed in the future; and a abnormal body temperature detection module for generating an abnormal body temperature detection result of the object under test based on the correlation coefficient.

[0050] Since the range of variation of body surface temperature always varies within a relatively small range, in order to be able to more accurately compare the difference between the time series of predicted body surface temperature and measured body surface temperature within a certain period of time, in some embodiments, the body temperature abnormality detection device 33 may include a fluctuation range amplification module for increasing the fluctuation range of the predicted body surface temperature after normalization and the measured body surface temperature observed in the future.

[0051] The body surface temperature prediction device 32 and the abnormal body temperature detection device 33 can be set on the server side, and can be a software module, a hardware module, or a combination of a software module and a hardware module.

[0052] The body surface temperature prediction device 32 constructed by a deep learning model based on the Temporal-Transformer framework will be taken as an example to describe and illustrate the body surface temperature prediction device 32.

[0053] Figure 4 : is a schematic diagram of the structure of the body surface temperature prediction device of this embodiment. Figure 4 As shown, the body surface temperature prediction device includes: a feature input part 40, a feature processing part 41 and a feature output part 42.

[0054] The feature input part 40 includes a static variable encoder 401 and a time-varying variable encoder 402 .

[0055] The static variable encoder 401 is used to encode the static variables of the measured object into a static feature vector, wherein the static variables include attribute information of the measured object, such as animal breed, age, geographical location of the farm, etc.

[0056] The time-varying variable encoder 402 is used to encode the first time-varying variable of the object under test observed in the past and the known second time-varying variable into a first time-varying feature vector, wherein the first time-varying variable includes the measured body surface temperature observed in the past, and the second time-varying variable includes the ambient temperature and time information, and the time information includes: quarter, month and time of day.

[0057] The time-varying feature vectors are encoded by embedding to retain the correlation between the time-varying feature vectors, or both the static feature vectors and the time-varying feature vectors are encoded by embedding to retain the correlation between the feature vectors.

[0058] Figure 4 The time-varying feature vector of each step in corresponds to a past observation point or a future prediction point. In this embodiment, the number of past observation points is multiple, and the number of future prediction points is one.

[0059] The feature processing part 41 includes a time series converter 411, which includes a static enhancement layer (StaticEnrichment Layer, referred to as SEL), a temporal self-attention layer (Temporal Self-Attention Layer, referred to as TSL) and a position-wise feed-forward network layer (Position-wise Feed-forward Layer, referred to as PFL). Among them, the static enhancement layer is used to introduce static features, and the temporal self-attention layer can learn the long-term dependencies of time series data and provide interpretability for the model. The position-wise feed-forward network layer is used to apply additional nonlinear processing to the output of the temporal self-attention layer.

[0060] In the model feature fusion solution, the related art usually concats all features and then extracts features. This method assumes that at each moment, all input features have the same impact on the predicted target, which is obviously inconsistent with the seasonal characteristics of the copper sheet of the ear tag sensor (the ambient temperature in autumn and winter has a greater impact on ear temperature, and the impact in spring and summer is smaller). In this embodiment, a gated residual network is used in the static enhancement layer to adjust the weights of the static feature vector and the first time-varying feature vector during feature fusion.

[0061] Among them, GLU is used in the Gated Residual Network (GRN) to control the degree of nonlinear contribution, where the formula of GLU is as follows: In the above formula, the input concatenated features X, W, b, V, c are the parameters that need to be learned. That is, after the input is mapped, it is weighted, and this weight is also normalized by the input mapping transformation. The result after sigmoid is between 0 and 1, so it can play a soft feature selection function. When some features are almost useless in the training process, the corresponding value after sigmoid will gradually become close to 0, and the weight of the ambient temperature in winter will increase.

[0062] Using the above-mentioned abnormal body temperature detection scheme based on the timing converter, the average error of the predicted body surface temperature in the next three hours compared with the measured body surface temperature is 0.45°C, which meets the reliability requirements of animal body temperature detection.

[0063] Through the above embodiment, starting from the change law of ear tag temperature when the animal has abnormal body temperature, combined with the influence of ambient temperature on ear tag temperature measurement, the trend law of ear tag temperature is analyzed to judge whether the animal has fever, so that real-time detection can be achieved, and real-time alarm can be issued for abnormalities in time. Animals that may have fever can be automatically detected in batches and alarms can be issued, which is convenient for farmers to conduct targeted re-inspections. In addition, the above solution adopts an end-to-end real-time detection process to ensure the availability of the service.

[0064] An embodiment of the present invention further provides an electronic device, comprising: at least one processor; and a memory in communication with the at least one processor. The memory stores a computer program executable by the at least one processor, and when the at least one processor executes the computer program, the computer program is used to enable the electronic device to execute the abnormal body temperature detection method of the embodiment of the present invention.

[0065] An embodiment of the present invention further provides a non-transitory machine-readable medium storing a computer program, wherein the computer program, when executed by a processor of a computer, is used to enable the computer to execute the abnormal body temperature detection method of an embodiment of the present invention.

[0066] An embodiment of the present invention further provides a computer program product, including a computer program, wherein the computer program, when executed by a processor of a computer, is used to enable the computer to execute the abnormal body temperature detection method of the embodiment of the present invention.

[0067] refer to Figure 5 , a block diagram of an electronic device that can be used as a server or client of an embodiment of the present invention will now be described, which is an example of a hardware device that can be applied to various aspects of the present invention. Electronic devices are intended to represent various forms of digital electronic computer devices, such as laptop computers, desktop computers, workbenches, personal digital assistants, servers, blade servers, mainframe computers, and other suitable computers. Electronic devices can also represent various forms of mobile devices, such as personal digital processing, cellular phones, smart phones, wearable devices, and other similar computing devices. The components shown herein, their connections and relationships, and their functions are merely examples and are not intended to limit the implementation of the present invention described and / or required herein.

[0068] like Figure 5As shown, the electronic device includes a computing unit 501, which can perform various appropriate actions and processes according to a computer program stored in a read-only memory (ROM) 502 or a computer program loaded from a storage unit 508 to a random access memory (RAM) 503. In the RAM 503, various programs and data required for the operation of the electronic device can also be stored. The computing unit 501, the ROM 502, and the RAM 503 are connected to each other via a bus 504. An input / output (I / O) interface 505 is also connected to the bus 504.

[0069] Multiple components in the electronic device are connected to the I / O interface 505, including: an input unit 506, an output unit 507, a storage unit 508, and a communication unit 509. The input unit 506 can be any type of device that can input information to the electronic device, and the input unit 506 can receive input digital or character information, and generate key signal input related to user settings and / or function control of the electronic device. The output unit 507 can be any type of device that can present information, and can include but is not limited to a display, a speaker, a video / audio output terminal, a vibrator, and / or a printer. The storage unit 508 can include but is not limited to a disk, an optical disk. The communication unit 509 allows the electronic device to exchange information / data with other devices through a computer network such as the Internet and / or various telecommunication networks, and can include but is not limited to a modem, a network card, an infrared communication device, a wireless communication transceiver, and / or a chipset, such as a Bluetooth device, a WiFi device, a WiMax device, a cellular communication device, and / or the like.

[0070] The computing unit 501 may be a variety of general and / or special processing components with processing and computing capabilities. Some examples of the computing unit 501 include, but are not limited to, a CPU, a graphics processing unit (GPU), various dedicated artificial intelligence (AI) computing chips, various computing units running machine learning model algorithms, digital signal processors (DSPs), and any appropriate processors, controllers, microcontrollers, etc. The computing unit 501 performs the various methods and processes described above. For example, in some embodiments, the method embodiments of the present invention may be implemented as a computer program, which is tangibly contained in a machine-readable medium, such as a storage unit 508. In some embodiments, part or all of the computer program may be loaded and / or installed on an electronic device via a ROM 502 and / or a communication unit 509. In some embodiments, the computing unit 501 may be configured to perform the above-described method in any other appropriate manner (e.g., by means of firmware).

[0071] The computer programs for implementing the methods of the embodiments of the present invention may be written in any combination of one or more programming languages. These computer programs may be provided to a processor or controller of a general-purpose computer, a special-purpose computer, or other programmable data processing device, so that when the computer programs are executed by the processor or controller, the functions / operations specified in the flow chart and / or block diagram are implemented. The computer programs may be executed entirely on the machine, partially on the machine, partially on the machine and partially on a remote machine as a stand-alone software package, or entirely on a remote machine or server.

[0072] In the context of an embodiment of the present invention, a machine-readable medium may be a tangible medium that may contain or store a program for use by or in conjunction with an instruction execution system, device, or equipment. A machine-readable medium may be a machine-readable signal medium or a machine-readable storage medium. A machine-readable signal medium may include, but is not limited to, an electronic, magnetic, optical, electromagnetic, infrared, or semiconductor system, device, or device, or any suitable combination of the foregoing. A more specific example of a machine-readable storage medium may include an electrical connection based on one or more lines, a portable computer disk, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), an optical fiber, a portable compact disk read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the foregoing.

[0073] It should be noted that the term "including" and its variations used in the embodiments of the present invention are open inclusions, that is, "including but not limited to". The term "based on" means "based at least in part on". The term "one embodiment" means "at least one embodiment"; the term "another embodiment" means "at least one other embodiment"; the term "some embodiments" means "at least some embodiments". The modifications of "one" and "multiple" mentioned in the embodiments of the present invention are illustrative rather than restrictive. Those skilled in the art should understand that unless the context clearly indicates otherwise, it should be understood as "one or more".

[0074] The user information (including but not limited to user device information, user personal information, etc.) and data (including but not limited to data used for analysis, stored data, displayed data, etc.) involved in the embodiments of the present invention are all information and data authorized by the user or fully authorized by all parties, and the collection, use and processing of relevant data must comply with the relevant laws, regulations and standards of relevant countries and regions, and provide corresponding operation entrances for users to choose to authorize or refuse.

[0075] The various steps described in the method implementation methods provided in the embodiments of the present invention may be performed in different orders and / or in parallel. In addition, the method implementation methods may include additional steps and / or omit the steps shown. The scope of protection of the present invention is not limited in this respect.

[0076] The term "embodiment" in this specification refers to specific features, structures or characteristics described in conjunction with the embodiment that can be included in at least one embodiment of the present invention. The appearance of this phrase in various places in the specification does not necessarily mean the same embodiment, nor does it mean that it is mutually exclusive with other embodiments and is independent or optional. The various embodiments in this specification are described in a related manner, and the same or similar parts between the various embodiments refer to each other. In particular, for the device, equipment, and system embodiments, since they are basically similar to the method embodiments, the description is relatively simple, and the relevant parts refer to the partial description of the method embodiment.

[0077] The above-mentioned embodiments only express several implementation methods of the present invention, and the description thereof is relatively specific and detailed, but it cannot be understood as limiting the scope of patent protection. It should be pointed out that, for a person of ordinary skill in the art, several variations and improvements can be made without departing from the concept of the present invention, and these all belong to the scope of protection of the present invention. Therefore, the scope of protection of the present invention shall be subject to the attached claims.

Claims

1. An abnormal body temperature detection system, comprising: A body surface temperature measuring device, used to measure the body surface temperature of the object being measured; An ambient temperature measuring device, used to measure the ambient temperature of the environment where the measured object is located; A body surface temperature prediction device, used to predict the predicted body surface temperature of the measured object in the future based on a first time-varying variable observed in the past and a known second time-varying variable of the measured object, wherein the first time-varying variable includes the measured body surface temperature observed in the past, and the second time-varying variable includes the ambient temperature and time information, and the time information includes: quarter, month and time of day; The abnormal body temperature detection device is used to detect the abnormal body temperature of the object under test based on the predicted body surface temperature and the measured body surface temperature observed in the future.

2. The system according to claim 1, wherein: The body surface temperature prediction device comprises: A static variable encoder, used for encoding a static variable into a static feature vector, wherein the static variable includes attribute information of the measured object; A time-varying variable encoder, used for encoding the first time-varying variable and the second time-varying variable into a first time-varying feature vector; A time series converter is used to process the static feature vector and the first time-varying feature vector based on pre-trained parameters to obtain a second time-varying feature vector for representing the predicted future body surface temperature of the measured object.

3. The system according to claim 2, wherein: The temporal converter includes: a gated residual network, which is used to adjust the weights of the static feature vector and the first time-varying feature vector during feature fusion.

4. The system according to claim 1, wherein: The abnormal body temperature detection device comprises: A mutual correlation coefficient calculation module, used to calculate the mutual correlation coefficient between the predicted body surface temperature and the measured body surface temperature observed in the future; The abnormal body temperature detection module is used to generate an abnormal body temperature detection result of the object under test based on the mutual correlation coefficient.

5. The system according to claim 4, wherein: The abnormal body temperature detection device also includes: The fluctuation range amplification module is used to increase the fluctuation range of the predicted body surface temperature after normalization and the measured body surface temperature observed in the future.

6. The system according to any one of claims 1 to 5, wherein: The time span of the predicted body surface temperature and the measured body surface temperature to be observed in the future ranges from one to three hours.

7. A body surface temperature prediction device, comprising: A static variable encoder, used for encoding static variables of the measured object into static feature vectors, wherein the static variables include attribute information of the measured object; A time-varying variable encoder, used for encoding a first time-varying variable observed in the past and a known second time-varying variable of the measured object into a first time-varying feature vector, wherein the first time-varying variable includes a measured body surface temperature observed in the past, and the second time-varying variable includes an ambient temperature and time information, and the time information includes: quarter, month and time of day; A time series converter is used to process the static feature vector and the first time-varying feature vector based on pre-trained parameters to obtain a second time-varying feature vector for representing the predicted future body surface temperature of the measured object.

8. The device according to claim 7, wherein: The temporal converter includes: a gated residual network, which is used to adjust the weights of the static feature vector and the first time-varying feature vector during feature fusion.

9. A method for detecting abnormal body temperature, comprising: Measure the body surface temperature of the object being measured and record the corresponding time information; Measuring the ambient temperature of the environment where the measured object is located and recording the corresponding time information; Predicting the predicted future body surface temperature of the measured object based on the measured body surface temperature observed in the past, as well as the known ambient temperature and time information, wherein the time information includes: quarter, month and time of day; Based on the predicted body surface temperature and the measured body surface temperature observed in the future, abnormal body temperature of the object being measured is detected.

10. The method according to claim 9, wherein: Detecting abnormal body temperature of the object under test based on the predicted body surface temperature and the measured body surface temperature observed in the future includes: Normalizing the predicted body surface temperature and the measured body surface temperature observed in the future, and increasing the fluctuation range of the processed body surface temperature; Calculating a correlation coefficient between the predicted body surface temperature and the measured body surface temperature observed in the future; Based on the mutual correlation coefficient, a body temperature abnormality detection result of the measured object is generated.

Images