Smart home equipment control method and system for home health care

Abstract

The invention provides a smart home equipment control method and system for home health, and the method comprises the steps: obtaining training samples, enabling each sample to comprise a user ID, a home ID, a code of a region where a user is located, temperature, humidity, a time period, and a human body existence state, carrying out the marking of each sample, and enabling the marking information to comprise the adjustment information of each piece of smart home equipment; constructing and training a health-care home model according to the labeled samples; and obtaining personal information of a current home person, identifying whether the current home person is a healthy person, if the current home person is the healthy person, monitoring current dynamic information of the healthy person, and combining the current time information, the current environment information and the healthy home model to obtain adjustment information of each smart home device so as to adjust the smart home device. The intelligent home equipment is automatically adjusted by combining the body information, time information and environmental factors of the elderly, the health of the elderly is actively protected, and therefore the life quality and safety of the elderly are improved.

Description

Technical Field

[0001] The present invention relates to the field of smart home technology, and in particular to a method and system for controlling smart home equipment for home health care. Background Art

[0002] Smart homes are widely used in all aspects of life, bringing a lot of convenience to daily life. At present, when using smart homes, smart homes often help users save operations based on their daily habits and automatically help users control smart home devices. However, in the health care system, many elderly people have poor autonomy, and the daily environmental conditions are different. It is easy to make misjudgments based on user habits, which in turn leads to errors in issuing execution instructions. Summary of the invention

[0003] The purpose of the present invention is to provide a smart home device control method and system for home health care to improve the above-mentioned problems.

[0004] In order to achieve the above objectives, the present application provides the following technical solutions:

[0005] On the one hand, an embodiment of the present application provides a method for controlling a smart home device for home health care, the method comprising:

[0006] Obtain training samples, each of which contains a user ID, a home ID, a user's region code, temperature, humidity, time period, and human presence status, and annotate each sample. The annotated information includes adjustment information of each smart home device.

[0007] Construct and train a health care home model based on the labeled samples;

[0008] Obtain the personal information of the current occupants of the home and identify whether they are health care personnel. If they are health care personnel, monitor their current dynamic information, and combine the current time information, current environmental information and health care home model to obtain the adjustment information of each smart home device in order to adjust the smart home devices.

[0009] In a second aspect, an embodiment of the present application provides a smart home device control system for home health care, the system comprising:

[0010] The acquisition module is used to obtain training samples. Each sample contains the user ID, home ID, user area code, temperature, humidity, time period, and human presence status. Each sample is labeled, and the labeled information includes the adjustment information of each smart home device.

[0011] The training module is used to build and train the health care home model based on the labeled samples;

[0012] The adjustment module is used to obtain the personal information of the current resident and identify whether they are health care personnel. If they are health care personnel, the adjustment module monitors the current dynamic information of the health care personnel and combines the current time information, current environmental information and health care home model to obtain the adjustment information of each smart home device in order to adjust the smart home devices.

[0013] In a third aspect, an embodiment of the present application provides a smart home device control device for home health care, the device comprising a memory and a processor. The memory is used to store a computer program; the processor is used to implement the steps of the smart home device control method for home health care when executing the computer program.

[0014] In a fourth aspect, an embodiment of the present application provides a readable storage medium, on which a computer program is stored. When the computer program is executed by a processor, the steps of the above-mentioned smart home device control method for home health care are implemented.

[0015] The beneficial effects of the present invention are:

[0016] 1. In the process of model training for identifying whether a person is a health care worker, the present invention screens the collected historical data for abnormalities, improves the accuracy of the samples, and also enriches the samples. By processing the samples, the accuracy of model recognition can be improved.

[0017] 2. The present invention combines the elderly's physical information, time information and environmental factors to automatically adjust smart home devices, actively protect the elderly's health, and thus improve the elderly's quality of life and safety.

[0018] Other features and advantages of the present invention will be described in the following description, and partly become apparent from the description, or be understood by implementing the embodiments of the present invention. The purpose and other advantages of the present invention can be realized and obtained by the structures particularly pointed out in the written description, claims, and drawings. BRIEF DESCRIPTION OF THE DRAWINGS

[0019] In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the drawings required for use in the embodiments are briefly introduced below. It should be understood that the following drawings only show certain embodiments of the present invention and therefore should not be regarded as limiting the scope. For ordinary technicians in this field, other related drawings can be obtained based on these drawings without creative work.

[0020] Figure 1 It is a flowchart of a method for controlling a smart home device for home health care described in an embodiment of the present invention;

[0021] Figure 2It is a schematic diagram of the structure of the smart home equipment control system for home health care described in an embodiment of the present invention;

[0022] Figure 3 It is a schematic diagram of the structure of the smart home device control device for home health care described in an embodiment of the present invention. DETAILED DESCRIPTION

[0023] In order to make the purpose, technical solutions and advantages of the embodiments of the present invention clearer, the technical solutions in the embodiments of the present invention will be clearly and completely described below in conjunction with the drawings in the embodiments of the present invention. Obviously, the described embodiments are part of the embodiments of the present invention, rather than all of the embodiments. The components of the embodiments of the present invention generally described and shown in the drawings here can be arranged and designed in various different configurations. Therefore, the following detailed description of the embodiments of the present invention provided in the drawings is not intended to limit the scope of the claimed invention, but merely represents selected embodiments of the present invention. Based on the embodiments in the present invention, all other embodiments obtained by ordinary technicians in this field without making creative work are within the scope of protection of the present invention.

[0024] It should be noted that similar reference numerals or letters represent similar items in the following drawings, so once an item is defined in one drawing, it does not need to be further defined and explained in the subsequent drawings. At the same time, in the description of the present invention, the terms "first", "second", etc. are only used to distinguish the description and cannot be understood as indicating or implying relative importance.

[0025] Example 1

[0026] like Figure 1 As shown, this embodiment provides a method for controlling smart home devices for home health care, which includes step S1, step S2 and step S3.

[0027] Step S1, obtaining training samples, each sample contains user ID, home ID, user area code, temperature, humidity, time period, human presence status, and annotating each sample, and the annotated information includes adjustment information of each smart home device;

[0028] In this step, the user ID and the home ID may be assigned by the staff to each user, and the region code of the user may be the code assigned by the staff to each region; the temperature and humidity are the current indoor temperature and humidity, which may be collected by the temperature and humidity sensor; the time period is the time period to which the current moment belongs, and the length of the time period may be customized, for example, the length of a time period may be one hour; the human presence status may be collected by the human presence sensor, and the human presence status includes presence / no one, activity / stillness; the annotation information may include the adjustment information of each smart home device, wherein the adjustment information of each smart home device further includes:

[0029] Air conditioning adjustment information: air conditioning temperature setting, switch, mode, wind speed;

[0030] Curtain adjustment information: curtain switch, opening and closing degree;

[0031] Door and window adjustment information: opening / closing of doors and windows;

[0032] Light adjustment information: light switch, brightness, and color temperature;

[0033] In addition to the above-mentioned smart home devices, it can also include mobile phones, cameras, etc.;

[0034] Step S2: construct and train a health care home model based on the labeled samples;

[0035] In this step, in terms of model selection, according to the pre-research algorithm, from decision forest, SVM support vector machine and KNN, the KNN algorithm is selected for training because it not only has better training effect, but also has fewer parameters than the decision forest model and is more suitable for deployment. Train the machine learning model and adjust the parameter configuration according to the results until the training results reach a relatively satisfactory prediction accuracy, and generate the final model.

[0036] Step S3, obtain the personal information of the current resident, identify whether he is a health care person, and if so, monitor the current dynamic information of the health care person, and combine the current time information, current environmental information and health care home model to obtain the adjustment information of each smart home device to adjust the smart home device.

[0037] In this step, if the person is a health care personnel, the current dynamic information of the health care personnel is monitored and combined with the current time information, current environmental information and the health care home model to obtain the adjustment information of each smart home device. Specifically, the user ID, home ID, user area code, temperature, humidity, time period, and human body status corresponding to the current home person are input into the health care home model, and the adjustment information of each smart home device is output; each smart home device is adjusted according to the adjustment information of these smart home devices.

[0038] At the same time, in this step, the personal information of the current home occupant is obtained, and the specific implementation steps of identifying whether the home occupant is a health care worker include step S31 and step S32;

[0039] Step S31, obtaining personal information of multiple historical home-based persons, saving the personal information of the historical home-based persons in an image format to obtain a first image, and collecting all the first images to obtain a first training set;

[0040] Step S32: perform abnormal screening on the first training set to obtain a second training set after screening; label each first image in the second training set, where different labels represent whether it is a health care personnel or not; train a machine learning model based on the labeled first image to obtain a health care personnel identification model, convert the personal information of the current home person into an image format, and input the converted information into the health care personnel identification model to obtain a result of whether the current home person is a health care personnel.

[0041] In this step, 1 can be used to represent a health care worker, and 2 can be used to represent a non-health care worker. The method in this step can delete all abnormal images in the first image, and this method can improve the accuracy of subsequent model training;

[0042] Meanwhile, in this step, the first training set is screened for abnormalities, and the specific implementation steps of obtaining the second training set after screening include step S321;

[0043] Step S321, perform cluster analysis on all the first images to form multiple clusters, and arbitrarily select two clusters from the multiple clusters to form a cluster pair; for each cluster pair, record the two clusters in the cluster pair as the first cluster and the second cluster, calculate the similarity between each first image in the first cluster and each first image in the second cluster, and obtain multiple similarity calculation results, count the number of similarity calculation results to obtain a first number, count the number of similarity calculation results that exceed a preset first similarity threshold to obtain a second number, calculate the ratio of the second number to the first number, and when the ratio is greater than the preset ratio threshold, delete the first image included in the cluster pair from the first training set to obtain a second training set.

[0044] Meanwhile, in this step, the specific implementation steps of training the machine learning model based on the annotated first image to obtain the health care personnel identification model include step S322;

[0045] Step S322, obtain a new image set, extract feature information of each new image and the first image remaining after screening; calculate the similarity between each new image and each first image based on the feature information, obtain multiple similarity calculation results, and use the new image corresponding to the smallest similarity in the similarity calculation results as the target image; input the target image and the annotated first image into the AdaIN model to obtain a new second image, the label of the second image is consistent with the label of the corresponding first image, and all the second images are grouped together to obtain a third training set; train the machine learning model based on the second training set and the third training set to obtain a health care personnel recognition model.

[0046] In this step, the new image is any image, and the second image is generated by the AdaIN model. The label information of the second image is consistent with the label information of the first image after annotation. This method can improve the richness of the sample. At the same time, the images in the second training set and the third training set can be input into the convolutional neural network model for training to obtain a health care personnel recognition model. In this step, the smaller the similarity calculation result, the greater the difference between the images, which can make the sample richer.

[0047] After step S3, the method further includes:

[0048] Step S4, obtaining the voice information sent by the current user, converting the voice information into text content, and obtaining the current text; obtaining a corresponding table, wherein the corresponding table includes multiple instruction information, and the instruction information includes a first text and a second text, wherein the first text includes environmental weather information and user demand information, and the second text is the adjustment information of the smart home device corresponding to the first text; performing similarity analysis on the current text and all the first texts according to formula (1), wherein formula (1) is:

[0049]

[0050] In formula (1), J represents the similarity between the current text and the first text, L represents the number of repeated words between the current text and the first text, T represents the total number of words in the first text, and a represents the logarithm base and is greater than 1;

[0051] In this step, the corresponding table is preset and stored in the system, which includes a large number of instruction information, wherein the first text can be understood as containing only environmental weather information and user demand information, the environmental weather information includes temperature, humidity, etc.; the user demand information includes adjusting to a comfortable temperature, comfortable humidity, etc.; the instruction information can be understood as: when the first text is "today's temperature is 36 degrees, adjust the indoor temperature to a comfortable temperature", the second text can be "adjust the air conditioning temperature to 25 degrees", and the voice information can also include environmental weather information and user demand information;

[0052] Step S5: extract the first text corresponding to the maximum similarity, and extract the corresponding second text, and display it on the mobile phone display screen to help the user adjust the smart home device.

[0053] In this step, corresponding adjustment information can be given according to the user's voice; after this step extracts the text with the greatest similarity, it can be sent to the user's mobile phone for the user to select, and the user can make corresponding adjustments according to the displayed adjustment information. At the same time, a display device can also be installed on the indoor wall and send it to the display device on the wall to facilitate elderly users to view.

[0054] Example 2

[0055] like Figure 3 As shown, this embodiment provides a smart home device control system for home health care, and the system includes an acquisition module 1, a training module 2 and an adjustment module 3.

[0056] Acquisition module 1 is used to acquire training samples, each sample contains user ID, home ID, user area code, temperature, humidity, time period, human presence status, and annotate each sample. The annotated information includes adjustment information of each smart home device.

[0057] Training module 2, used to build and train a health care home model based on the labeled samples;

[0058] Adjustment module 3 is used to obtain the personal information of the current resident and identify whether they are health care personnel. If they are health care personnel, the adjustment module 3 monitors the current dynamic information of the health care personnel and combines the current time information, current environmental information and health care home model to obtain the adjustment information of each smart home device to adjust the smart home device.

[0059] In a specific implementation of the present disclosure, the adjustment module 3 further includes a first acquisition unit 31 and a training unit 32 .

[0060] The acquisition unit 31 is used to acquire the personal information of multiple historical home-based persons, save the personal information of the historical home-based persons in the format of an image, obtain a first image, and collect all the first images to obtain a first training set;

[0061] The training unit 32 is used to perform abnormal screening on the first training set to obtain a second training set after screening; label each first image in the second training set, where different labels represent whether it is a health care personnel or not; train a machine learning model based on the labeled first image to obtain a health care personnel identification model, convert the personal information of the current home person into an image format, and input the converted personal information into the health care personnel identification model to obtain a result of whether the current home person is a health care personnel.

[0062] In a specific implementation of the present disclosure, the training unit 32 further includes a screening unit 321 .

[0063] The screening unit 321 is used to perform cluster analysis on all the first images to form multiple clusters, and arbitrarily select two clusters from the multiple clusters to form a cluster pair; for each cluster pair, the two clusters in the cluster pair are respectively recorded as the first cluster and the second cluster, and the similarity between each first image in the first cluster and each first image in the second cluster is calculated to obtain multiple similarity calculation results, and the number of similarity calculation results is counted to obtain a first number, and the number of similarity calculation results exceeding a preset first similarity threshold is counted to obtain a second number, and the ratio of the second number to the first number is calculated. When the ratio is greater than the preset ratio threshold, the first image included in the cluster pair is deleted from the first training set to obtain a second training set.

[0064] In a specific implementation of the present disclosure, the training unit 32 further includes an acquisition unit 322 .

[0065] The acquisition unit 322 is used to acquire a new image set, extract feature information of each new image and the first image remaining after screening; calculate the similarity between each new image and each first image based on the feature information, obtain multiple similarity calculation results, and use the new image corresponding to the smallest similarity in the similarity calculation results as the target image; input the target image and the annotated first image into the AdaIN model to obtain a new second image, the label of the second image is consistent with the label of the corresponding first image, and all the second images are collected to obtain a third training set; train the machine learning model based on the second training set and the third training set to obtain a health care personnel recognition model.

[0066] In a specific embodiment of the present disclosure, the system further includes a conversion module 4 and a display module 5.

[0067] The conversion module 4 is used to obtain the voice information sent by the current user, convert the voice information into text content, and obtain the current text; obtain a corresponding table, the corresponding table includes multiple instruction information, the instruction information includes a first text and a second text, the first text includes environmental weather information and user demand information, and the second text is the adjustment information of the smart home device corresponding to the first text; perform similarity analysis on the current text and all the first texts according to formula (1), and the formula (1) is:

[0068]

[0069] In formula (1), J represents the similarity between the current text and the first text, L represents the number of repeated words between the current text and the first text, T represents the total number of words in the first text, and a represents the logarithm base and is greater than 1;

[0070] The display module 5 is used to extract the first text corresponding to the maximum similarity, and extract the corresponding second text, and display it on the mobile phone display screen to help the user adjust the smart home device.

[0071] It should be noted that, regarding the system in the above embodiment, the specific manner in which each module performs operations has been described in detail in the embodiment of the method, and will not be elaborated here.

[0072] Example 3

[0073] Corresponding to the above method embodiments, the embodiments of the present disclosure also provide a smart home device control device for home health care. The smart home device control device for home health care described below and the smart home device control method for home health care described above can be referenced to each other.

[0074] Figure 3 FIG. 3 is a block diagram of a smart home device control device 300 for home health care according to an exemplary embodiment. Figure 3 As shown, the smart home device control device 300 for home health care may include: a processor 301, a memory 302. The smart home device control device 300 for home health care may also include one or more of a multimedia component 303, an I / O interface 304, and a communication component 305.

[0075] The processor 301 is used to control the overall operation of the smart home device control device 300 for home health care to complete all or part of the steps in the smart home device control method for home health care. The memory 302 is used to store various types of data to support the operation of the smart home device control device 300 for home health care, which may include, for example, instructions for any application or method operating on the smart home device control device 300 for home health care, as well as application-related data, such as contact data, messages sent and received, pictures, audio, video, etc. The memory 302 can be implemented by any type of volatile or non-volatile storage device or a combination thereof, such as static random access memory (SRAM), electrically erasable programmable read-only memory (EEPROM), erasable programmable read-only memory (EPROM), programmable read-only memory (PROM), read-only memory (ROM), magnetic memory, flash memory, disk or optical disk. The multimedia component 303 may include a screen and an audio component. The screen may be, for example, a touch screen, and the audio component is used to output and / or input audio signals. For example, the audio component may include a microphone for receiving external audio signals. The received audio signal may be further stored in the memory 302 or sent via the communication component 305. The audio component also includes at least one speaker for outputting audio signals. The I / O interface 304 provides an interface between the processor 301 and other interface modules, and the above-mentioned other interface modules can be keyboards, mice, buttons, etc. These buttons can be virtual buttons or physical buttons. The communication component 305 is used for wired or wireless communication between the smart home device control device 300 for home health care and other devices. Wireless communication, such as Wi-Fi, Bluetooth, Near Field Communication (NFC), 2G, 3G or 4G, or a combination of one or more of them, so the corresponding communication component 305 can include: Wi-Fi module, Bluetooth module, NFC module.

[0076] In an exemplary embodiment, the smart home device control device 300 for home health care can be implemented by one or more application specific integrated circuits (Application Specific Integrated Circuit, referred to as ASIC), digital signal processors (Digital Signal Processing Device, referred to as DSP), digital signal processing devices (Digital Signal Processing Device, referred to as DSPD), programmable logic devices (Programmable Logic Device, referred to as PLD), field programmable gate arrays (Field Programmable Gate Array, referred to as FPGA), controllers, microcontrollers, microprocessors or other electronic components to execute the above-mentioned smart home device control method for home health care.

[0077] In another exemplary embodiment, a computer-readable storage medium including program instructions is also provided, and when the program instructions are executed by a processor, the steps of the above-mentioned smart home device control method for home health care are implemented. For example, the computer-readable storage medium can be the above-mentioned memory 302 including program instructions, and the above-mentioned program instructions can be executed by the processor 301 of the smart home device control device 300 for home health care to complete the above-mentioned smart home device control method for home health care.

[0078] Example 4

[0079] Corresponding to the above method embodiments, the embodiments of the present disclosure also provide a readable storage medium. The readable storage medium described below and the smart home device control method for home health care described above can be referenced to each other.

[0080] A readable storage medium having a computer program stored thereon, which, when executed by a processor, implements the steps of the method for controlling smart home devices for home health care of the above-mentioned method embodiment.

[0081] The readable storage medium may specifically be a U disk, a mobile hard disk, a read-only memory (ROM), a random access memory (RAM), a magnetic disk or an optical disk, or other readable storage medium that can store program codes.

[0082] The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. For those skilled in the art, the present invention may have various modifications and variations. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present invention shall be included in the protection scope of the present invention.

Claims

1. A method for controlling smart home devices for home health care, characterized in that: include: Obtain training samples, each of which contains a user ID, a home ID, a user's region code, temperature, humidity, time period, and human presence status, and annotate each sample. The annotated information includes adjustment information of each smart home device. Construct and train a health care home model based on the labeled samples; Obtain the personal information of the current occupants of the home and identify whether they are health care personnel. If they are health care personnel, monitor their current dynamic information, and combine the current time information, current environmental information and health care home model to obtain the adjustment information of each smart home device in order to adjust the smart home devices.

2. The method for controlling smart home devices for home health care according to claim 1, characterized in that: Obtain the personal information of the current home occupants and identify whether they are health care personnel, including: Obtaining personal information of multiple historical home-based persons, saving the personal information of the historical home-based persons in an image format to obtain a first image, and collecting all the first images to obtain a first training set; The first training set is screened for abnormalities to obtain a second training set; each first image in the second training set is labeled, and different labels represent whether it is a health care personnel or not; a machine learning model is trained based on the labeled first images to obtain a health care personnel identification model, and the personal information of the current home person is converted into an image format, which is input into the health care personnel identification model after conversion to obtain a result of whether the current home person is a health care personnel.

3. The method for controlling smart home devices for home health care according to claim 2, characterized in that: The first training set is screened for abnormalities, and a second training set is obtained after the screening, including: Perform cluster analysis on all first images to form multiple clusters, and arbitrarily select two clusters from the multiple clusters to form a cluster pair; for each cluster pair, record the two clusters in the cluster pair as the first cluster and the second cluster, calculate the similarity between each first image in the first cluster and each first image in the second cluster, and obtain multiple similarity calculation results, count the number of similarity calculation results to obtain a first number, count the number of similarity calculation results that exceed a preset first similarity threshold to obtain a second number, calculate the ratio of the second number to the first number, and when the ratio is greater than the preset ratio threshold, delete the first image included in the cluster pair from the first training set to obtain a second training set.

4. The method for controlling smart home devices for home health care according to claim 2, characterized in that: The machine learning model is trained based on the annotated first image to obtain a health care personnel recognition model, including: A new image set is obtained, and feature information of each new image and the first images remaining after screening is extracted; based on the feature information, the similarity between each new image and each first image is calculated to obtain multiple similarity calculation results, and the new image corresponding to the smallest similarity in the similarity calculation results is used as the target image; the target image and the annotated first image are input into the AdaIN model to obtain a new second image, the label of the second image is consistent with the label of the corresponding first image, and all the second images are collected to obtain a third training set; a machine learning model is trained based on the second training set and the third training set to obtain a health care personnel recognition model.

5. The method for controlling smart home devices for home health care according to claim 4, characterized in that: Combining the current time information, the current environment information and the health care home model, the adjustment information of each smart home device is obtained to adjust the smart home device, and further includes: Acquire the voice information sent by the current user, convert the voice information into text content, and obtain the current text; acquire a corresponding table, the corresponding table includes multiple instruction information, the instruction information includes a first text and a second text, the first text includes environmental weather information and user demand information, and the second text is the adjustment information of the smart home device corresponding to the first text; perform similarity analysis on the current text and all the first texts according to formula (1), and the formula (1) is: In formula (1), J represents the similarity between the current text and the first text, L represents the number of repeated words between the current text and the first text, T represents the total number of words in the first text, and a represents the logarithm base and is greater than 1; The first text corresponding to the maximum similarity is extracted, and the corresponding second text is extracted and displayed on the mobile phone display screen to help the user adjust the smart home device.

6. A smart home equipment control system for home health care, characterized in that: include: The acquisition module is used to obtain training samples. Each sample contains the user ID, home ID, user area code, temperature, humidity, time period, and human presence status. Each sample is labeled, and the labeled information includes the adjustment information of each smart home device. The training module is used to build and train the health care home model based on the labeled samples; The adjustment module is used to obtain the personal information of the current resident and identify whether they are health care personnel. If they are health care personnel, the adjustment module monitors the current dynamic information of the health care personnel and combines the current time information, current environmental information and health care home model to obtain the adjustment information of each smart home device in order to adjust the smart home devices.

7. The smart home equipment control system for home health care according to claim 6, characterized in that: Regulation module, including: An acquisition unit, used for acquiring personal information of a plurality of historical home-based persons, saving the personal information of the historical home-based persons in an image format to obtain a first image, and collecting all the first images to obtain a first training set; The training unit is used to screen the first training set for abnormalities and obtain a second training set after screening; label each first image in the second training set, where different labels represent whether it is a health care personnel or not; train a machine learning model based on the labeled first image to obtain a health care personnel identification model, convert the personal information of the current home person into an image format, and input the converted personal information into the health care personnel identification model to obtain a result of whether the current home person is a health care personnel.

8. The smart home equipment control system for home health care according to claim 7, characterized in that: Training modules include: A screening unit is used to perform cluster analysis on all first images to form multiple clusters, and arbitrarily select two clusters from the multiple clusters to form a cluster pair; for each cluster pair, the two clusters in the cluster pair are respectively recorded as a first cluster and a second cluster, and the similarity between each first image in the first cluster and each first image in the second cluster is calculated to obtain multiple similarity calculation results, and the number of similarity calculation results is counted to obtain a first number, and the number of similarity calculation results exceeding a preset first similarity threshold is counted to obtain a second number, and the ratio of the second number to the first number is calculated. When the ratio is greater than the preset ratio threshold, the first image included in the cluster pair is deleted from the first training set to obtain a second training set.

9. The smart home equipment control system for home health care according to claim 7, characterized in that: Training modules include: An acquisition unit is used to acquire a new image set, extract feature information of each new image and the first image remaining after screening; calculate the similarity between each new image and each first image according to the feature information, obtain multiple similarity calculation results, and use the new image corresponding to the smallest similarity in the similarity calculation results as the target image; input the target image and the annotated first image into the AdaIN model to obtain a new second image, the label of the second image is consistent with the label of the corresponding first image, and all the second images are collected to obtain a third training set; train a machine learning model based on the second training set and the third training set to obtain a health care personnel recognition model.

10. The smart home equipment control system for home health care according to claim 9, characterized in that: Also includes: The conversion module is used to obtain the voice information sent by the current user, convert the voice information into text content, and obtain the current text; Obtain a correspondence table, the correspondence table including multiple instruction information, the instruction information including a first text and a second text, the first text including environmental weather information and user demand information, the second text is the adjustment information of the smart home device corresponding to the first text; perform similarity analysis on the current text and all the first texts according to formula (1), the formula (1) is: In formula (1), J represents the similarity between the current text and the first text, L represents the number of repeated words between the current text and the first text, T represents the total number of words in the first text, and a represents the logarithm base and is greater than 1; The display module is used to extract the first text corresponding to the maximum similarity, and extract the corresponding second text, and display it on the mobile phone display screen to help users adjust the smart home devices.

Images