Pension institution remote monitoring system based on image processing

Abstract

The invention, which relates to the technical field of intelligent monitoring, discloses an image-processing-based remote monitoring system for an old-age care institution, comprising an acquisition module, a first analysis module and a second analysis module, obtaining a first face image, constructing an emotion influence degree model and a sleep quality influence degree model, comparing a diet name with a diet suggestion, and obtaining a diet suggestion result; and obtaining a first comparison result, and executing a first operation according to the first comparison result. By identifying the service code of the to-be-tested object, quickly positioning and responding to the needs of the elderly, constructing an emotion influence degree model, providing more personalized care and support, constructing a sleep quality influence degree model and generating an environment regulation decision, the sleep quality is favorably improved, the diet of the elderly is ensured to accord with the doctor's advice, the health is promoted, and the health of the elderly is improved. The intelligent health equipment can also realize health file management of the old, know the change trend of the health condition of the old and comprehensively evaluate the health condition of the old.

Description

Technical Field

[0001] The present invention relates to the technical field of intelligent monitoring, and in particular to a remote monitoring system for a nursing home based on image processing. Background Art

[0002] In recent years, with the aging of the global population and the increase in the number of patients with chronic diseases, the demand for remote health monitoring has continued to rise. Technological progress has provided support for the remote health monitoring industry. The development of technologies such as the Internet of Things, big data, cloud computing and artificial intelligence has made monitoring equipment more intelligent and data analysis capabilities stronger.

[0003] At present, a Chinese invention patent with publication number CN112150114A discloses a smart elderly care system and construction method based on BIM technology. This method implements multi-dimensional information collection and display by three-dimensionally deploying the nursing home buildings and equipment on the management platform of the BIM model, thereby improving the management level of the nursing home. However, the relevant technology does not make constructive suggestions on the frequency of relatives' visits based on the actual psychological needs of the elderly, lacks humanistic care for the elderly, and does not scientifically monitor the sleep and diet of the elderly based on objective monitoring data. The monitoring lacks comprehensiveness, is not conducive to the physical and mental health of the elderly, and has certain limitations. Summary of the invention

[0004] The technical problem solved by the present invention is that the relevant technology does not make constructive suggestions on the frequency of relatives' visits based on the actual psychological needs of the elderly, lacks humanistic care for the elderly, does not conduct scientific monitoring of the elderly's sleep and diet based on objective monitoring data of the elderly, lacks comprehensiveness of monitoring, is not conducive to the physical and mental health of the elderly, and has certain limitations.

[0005] In order to solve the above technical problems, the present invention provides the following technical solutions: a remote monitoring system for a nursing home based on image processing, comprising a collection module, a first analysis module and a second analysis module;

[0006] The acquisition module records the service object's own facial image and related facial images, identifies the service code of the subject to be tested, matches the self-monitoring data, environmental data and dietary pictures corresponding to the service code of the subject to be tested, numbers the related facial images, and obtains a first facial image;

[0007] The first analysis module divides the monitoring time into a first time period and a second time period according to the brain wave data, obtains the self-monitoring data when the same first facial image appears in the first time period, recorded as the first data, calculates the interval length of adjacent first data in the first time period, recorded as the first duration, constructs an emotional impact model of the first facial image on the service object according to the first data and the first duration, generates a corresponding relative visit decision according to the emotional impact model, determines the sleep quality level according to the self-monitoring data in the second time period, constructs a sleep quality impact model according to the sleep quality level and environmental data, and generates an environmental control decision according to the sleep quality impact model;

[0008] The second analysis module obtains the doctor's advice information of the service object, identifies the name of the diet according to the diet picture, compares the diet name with the diet suggestion to obtain a first comparison result, and performs a first operation according to the first comparison result.

[0009] As a preferred solution of the remote monitoring system of a nursing home based on image processing described in the present invention, wherein: the acquisition module records the service object's own facial image and related facial images, the related facial images are represented as facial images of the service object's relatives, and identifies the service code of the object to be tested, the service code of the object to be tested is sensed by radio frequency identification through a bracelet worn by the service object, and the recording logic of the own facial image and related facial images includes;

[0010] Starting a face recording device, setting shooting parameters, wherein the shooting parameters include a shooting angle and a shooting focal length, sending a recording start signal, shooting a face, obtaining a captured image, identifying a face contour in the captured image, counting the number of pixels within the face contour, recorded as a first number, counting the total number of pixels in the captured image, recorded as a second number, calculating a ratio of the first number to the second number, recorded as a first ratio, wherein the first ratio is expressed as a proportion of valid pixels in the captured image, setting a first value as a threshold of the first ratio, and comparing the first ratio with the first value;

[0011] When the first ratio is greater than or equal to the first value, the captured image is bound to the service object code, and the label of the captured image is set to the relative name of the service object;

[0012] When the first ratio is less than the first value, the first gradient value is set as the change in the shooting angle, the second gradient value is set as the change in the shooting focal length, the shooting parameters are continuously adjusted according to the first gradient value and the second gradient value, and the adjusted first ratio is continuously calculated until the adjusted first ratio is greater than or equal to the first value, then the continuous adjustment of the shooting parameters is stopped, the adjusted shooting image is bound to the service object code, and the label of the adjusted shooting image is set to the relative name of the service object;

[0013] The continuous control is represented by continuous increase or continuous decrease, and the first gradient value and the second gradient value are obtained through big data;

[0014] When the captured image is bound to the service object code and the label of the captured image is set, the face entry device is closed;

[0015] The related facial images are numbered to obtain a first facial image, and the number of the related facial image is represented by X i , where i is a natural number, X i Represents the relatives of the service object whose face image is entered.

[0016] As a preferred solution of the remote monitoring system of a nursing home based on image processing described in the present invention, wherein: the acquisition module retrieves the service database, inputs the service object code into the service database, and matches the self-monitoring data, environmental data and dietary pictures corresponding to the service code of the object to be tested;

[0017] The self-monitoring data is brain wave data, and the brain wave data is represented by a brain wave image within a monitoring period;

[0018] The environmental data includes environmental noise, environmental humidity and environmental temperature;

[0019] The food pictures represent food pictures taken for lunch and food pictures taken for dinner.

[0020] As a preferred solution of the remote monitoring system for nursing homes based on image processing described in the present invention, wherein: the first analysis module divides the monitoring time into a first time period and a second time period according to the brain wave data;

[0021] The division logic of the first time period and the second time period includes:

[0022] The second value is set as the individual time step, and the third value is set as the time period, and the original brain wave signal is divided into sub-segments, the second value and the third value are obtained through big data, the second value represents the number of brain wave amplitudes contained in each sub-segment, and the third value represents the duration of each sub-segment, and the brain wave amplitudes in the sub-segments are obtained in time sequence, and the brain wave amplitudes are subjected to discrete wavelet transform to obtain the frequency domain expression of the amplitude, and the spectrum entropy value of the amplitude is calculated, and the average value of the spectrum entropy value of the amplitude is calculated and recorded as the first average value, and the standard spectrum entropy of the sleep period is obtained, and the first average value is compared with the standard spectrum entropy of the sleep period. When the first average value is greater than the standard spectrum entropy of the sleep period, the corresponding sub-segment is classified into the first time period, and when the first average value is less than or equal to the standard spectrum entropy of the sleep period, the corresponding sub-segment is classified into the second time period;

[0023] The first time point is set as the evening demarcation point, wherein the first time point is a generally recognized decomposition time point of evening and daytime, the sub-segments classified into the first time period and distributed before the first time point are set as the first time period, and the sub-segments classified into the second time period and distributed after the first time point are set as the second time period;

[0024] The first time period is represented as a time period of wakefulness, and the second time period is represented as a time period of sleep.

[0025] As a preferred solution of the remote monitoring system of a nursing home based on image processing described in the present invention, wherein: the first analysis module obtains brain wave data when the same first facial image appears in a first time period, recorded as first data, and counts the interval length of adjacent first data in the first time period, recorded as first duration, wherein the adjacent is represented by adjacent in time order;

[0026] According to the first data and the first duration, an emotion impact model of the first facial image on the service object is constructed, and a corresponding relative visit decision is generated according to the emotion impact model. The construction logic of the emotion impact model includes:

[0027] Acquire first data, set a first segmentation step length, wherein the first segmentation step length is obtained through historical experience, segment the first data to obtain first sub-data, perform band-pass filtering on the first sub-data, extract the frequency characteristic quantity of the first sub-data after band-pass filtering, record it as the first characteristic quantity, obtain the frequency characteristic quantity of the standard α wave and the frequency characteristic quantity of the standard β wave, record it as the second characteristic quantity and the third characteristic quantity, set the third value as the similarity threshold, calculate the similarity between the first characteristic quantity and the second characteristic quantity through the cosine similarity formula, record it as the first similarity, and calculate the similarity between the first characteristic quantity and the third characteristic quantity, record it as the second similarity, and compare the second similarity with the first similarity;

[0028] When the first similarity is greater than or equal to the third value, the first sub-data is set to the α wave; when the second similarity is greater than or equal to the third value, the first sub-data is set to the β wave; otherwise, the first sub-data is deleted;

[0029] The number of α wave appearances and the number of β wave appearances are counted respectively, recorded as the third number and the fourth number, the product of the third number and the segmentation step length, and the product of the fourth number and the segmentation step length are calculated respectively, recorded as the first product and the second product, the ratio of the second product to the first product is calculated, recorded as the second ratio, the second ratio represents the degree of excitement, and the larger the second ratio is, the more excited the person is;

[0030] Taking the second ratio as the dependent variable and the first duration as the independent variable, the emotional impact model was constructed through mathematical tools;

[0031] Set an excitement interval, obtain an upper limit value of the first duration and a lower limit value of the first duration according to the excitement interval, and obtain a corresponding relative visit decision according to the upper limit value of the first duration and the lower limit value of the first duration, wherein the corresponding relative visit decision includes a number X i For relatives, the minimum visit interval is the lower limit of the first duration, and the maximum visit interval is the upper limit of the first duration.

[0032] As a preferred solution of the remote monitoring system of a nursing home based on image processing described in the present invention, the calculation expression of the emotion impact model is:

[0033]

[0034] in, Expressed as The second ratio corresponding to the relatives, Represents the first duration, The transformation rule expressed as the ratio of the first duration to the second is obtained by mathematical tools.

[0035] As a preferred solution of the remote monitoring system for nursing homes based on image processing described in the present invention, the first analysis module determines the sleep quality level according to the brain wave data in the second time period, and the judgment logic of the sleep quality level includes:

[0036] Obtain brain wave data in a second time period, set a second segmentation step, the second segmentation step is obtained through historical experience, the brain wave data in the second time period is segmented to obtain second sub-data, the α wave power and the β wave power in each second sub-data are obtained, the average value of the α wave power and the average value of the β wave power are calculated respectively, which are recorded as the second average value and the third average value respectively, the ratio of the third average value to the second average value is calculated, which is recorded as the third ratio, the fourth value and the fifth value are set as the threshold of the third ratio, the fourth value and the fifth value are obtained through historical experience, the third ratio is compared with the fourth value and the fifth value, when the third ratio is less than the fourth value, the sleep quality level is set to the first level, when the third ratio is greater than or equal to the fourth value and less than the fifth value, the sleep quality level is set to the second level, when the third ratio is greater than or equal to the fifth value, the sleep quality level is set to the third level, and the sleep quality represented by the first level, the second level and the third level gradually decreases;

[0037] A sleep quality impact model is constructed according to the sleep quality level, environmental noise, environmental humidity and environmental temperature, and an environmental control decision is generated according to the sleep quality impact model.

[0038] As a preferred solution of the remote monitoring system for nursing homes based on image processing described in the present invention, the construction logic of the sleep quality impact model includes:

[0039] Taking the sleep quality level as the dependent variable and the ambient noise, ambient humidity and ambient temperature as the independent variables, a multiple regression analysis was performed to obtain a sleep quality impact model. The calculation expression of the sleep quality impact model is:

[0040]

[0041] in, is the sleep quality level, For the independent variables, For the The coefficients corresponding to the independent variables;

[0042] The current sleep quality level is obtained according to the sleep quality impact model. When the current sleep quality level is not the first level, the third gradient value and the fourth gradient value are set as the change in ambient humidity and the change in ambient temperature respectively, and the ambient temperature interval and the ambient humidity interval are set. Within the ambient temperature interval and the ambient humidity interval, the ambient humidity and the ambient temperature are continuously regulated according to the third gradient value and the fourth gradient value, and the current sleep quality level after regulation is continuously obtained. When the current sleep quality level is the first level, the continuous regulation of the ambient humidity and the ambient temperature is stopped. Otherwise, a sound insulation device is added, and the sound insulation device includes a sound insulation board or a sound insulation green belt. The width and height of the sound insulation device are continuously changed, and the current sleep quality level is changed to the first level.

[0043] As a preferred solution of the remote monitoring system of a nursing home based on image processing described in the present invention, the second analysis module is connected to the hospital system, inputs the service code of the service object into the hospital system, obtains the medical advice information of the service object, the medical advice information is expressed as dietary advice, and the dietary advice is expressed as the names of foods that cannot be eaten.

[0044] As a preferred solution of the remote monitoring system for nursing homes based on image processing described in the present invention, wherein: identifying food names according to dietary pictures, comparing the food names with dietary recommendations to obtain a first comparison result, performing a first operation according to the first comparison result, the first comparison result including that the food names comply with the dietary recommendations, and that the food names do not comply with the dietary recommendations, the first operation including not sending a food warning signal, and sending a food warning signal;

[0045] When the first comparison result is that the food name does not conform to the dietary recommendations, the first operation is set to send a food warning signal; otherwise, the first operation is set to not send a food warning signal.

[0046] The beneficial effects of the present invention are as follows: by collecting facial images and related facial images of the service object, the service code of the object to be tested can be identified, and the corresponding monitoring data can be matched, which is helpful to quickly locate and respond to the needs of the elderly and improve monitoring efficiency; the monitoring time is divided into different time periods according to the brain wave data, and an emotional impact model is constructed, which is helpful to monitor the emotional state of the elderly and generate corresponding relative visit decisions according to the model, thereby providing more personalized care and support; a sleep quality impact model can be constructed according to the sleep quality level and environmental data, and environmental regulation decisions can be generated, which is helpful to improve the sleeping environment of the elderly and improve sleep quality; the medical advice information of the service object can be obtained, and the name of the diet can be identified according to the diet picture, and compared with the diet recommendation to ensure that the elderly's diet is in line with the doctor's advice, which helps the elderly to better follow the doctor's advice and promote health; the smart health equipment can also realize the health record management of the elderly, understand the changing trend of the elderly's health status, and comprehensively evaluate the health status of the elderly. BRIEF DESCRIPTION OF THE DRAWINGS

[0047] Figure 1 A basic flow chart of a remote monitoring system for a nursing home based on image processing provided by one embodiment of the present invention. DETAILED DESCRIPTION

[0048] In order to make the above-mentioned objects, features and advantages of the present invention more obvious and easy to understand, the specific implementation methods of the present invention are described in detail below in conjunction with the drawings. It is obvious that the described embodiments are only part of the embodiments of the present invention, but not all of the embodiments.

[0049] Example, see Figure 1 , as an embodiment of the present invention, provides a remote monitoring system for a nursing home based on image processing, including a collection module, a first analysis module and a second analysis module;

[0050] The acquisition module records the service object's own facial image and related facial images, identifies the service code of the subject to be tested, matches the self-monitoring data, environmental data and dietary pictures corresponding to the service code of the subject to be tested, numbers the related facial images, and obtains a first facial image;

[0051] The first analysis module divides the monitoring time into a first time period and a second time period according to the brain wave data, obtains the self-monitoring data when the same first facial image appears in the first time period, recorded as the first data, calculates the interval length of adjacent first data in the first time period, recorded as the first duration, constructs an emotional impact model of the first facial image on the service object according to the first data and the first duration, generates a corresponding relative visit decision according to the emotional impact model, determines the sleep quality level according to the self-monitoring data in the second time period, constructs a sleep quality impact model according to the sleep quality level and environmental data, and generates an environmental control decision according to the sleep quality impact model;

[0052] The second analysis module obtains the doctor's advice information of the service object, identifies the name of the diet according to the diet picture, compares the diet name with the diet suggestion to obtain a first comparison result, and performs a first operation according to the first comparison result.

[0053] By collecting facial images and related facial images of the service object, the present invention can identify the service code of the object to be tested and match the corresponding monitoring data, which is helpful to quickly locate and respond to the needs of the elderly and improve monitoring efficiency. The monitoring time is divided into different time periods according to the brain wave data, and an emotional impact model is constructed, which is helpful to monitor the emotional state of the elderly and generate corresponding relative visit decisions according to the model, so as to provide more personalized care and support. The sleep quality impact model can be constructed according to the sleep quality level and environmental data, and environmental regulation decisions can be generated, which is helpful to improve the sleeping environment of the elderly and improve the sleep quality. The medical advice information of the service object is obtained, and the name of the diet is identified according to the diet picture, and compared with the diet suggestion to ensure that the diet of the elderly complies with the doctor's advice, which helps the elderly to better follow the doctor's advice and promote health. The smart health equipment can also realize the health record management of the elderly, understand the changing trend of the health status of the elderly, and comprehensively evaluate the health status of the elderly.

[0054] The acquisition module records the service object's own facial image and related facial images, the related facial images are facial images of the service object's relatives, and identifies the service code of the object to be tested, the service code of the object to be tested is sensed by radio frequency identification through the service object wearing a bracelet, and the recording logic of the own facial image and related facial images includes:

[0055] Starting a face recording device, setting shooting parameters, wherein the shooting parameters include a shooting angle and a shooting focal length, sending a recording start signal, shooting a face, obtaining a captured image, identifying a face contour in the captured image, counting the number of pixels within the face contour, recorded as a first number, counting the total number of pixels in the captured image, recorded as a second number, calculating a ratio of the first number to the second number, recorded as a first ratio, wherein the first ratio is expressed as a proportion of valid pixels in the captured image, setting a first value as a threshold of the first ratio, and comparing the first ratio with the first value;

[0056] When the first ratio is greater than or equal to the first value, the captured image is bound to the service object code, and the label of the captured image is set to the relative name of the service object;

[0057] When the first ratio is less than the first value, the first gradient value is set as the change in the shooting angle, the second gradient value is set as the change in the shooting focal length, the shooting parameters are continuously adjusted according to the first gradient value and the second gradient value, and the adjusted first ratio is continuously calculated until the adjusted first ratio is greater than or equal to the first value, then the continuous adjustment of the shooting parameters is stopped, the adjusted shooting image is bound to the service object code, and the label of the adjusted shooting image is set to the relative name of the service object;

[0058] The continuous control is represented by continuous increase or continuous decrease, and the first gradient value and the second gradient value are obtained through big data;

[0059] When the captured image is bound to the service object code and the label of the captured image is set, the face entry device is closed;

[0060] The related facial images are numbered to obtain a first facial image. The numbers of the related facial images are represented by Xi, where i is a natural number and Xi represents a relative of the service object whose i-th facial image is entered.

[0061] The acquisition module retrieves the service database, inputs the service object code into the service database, and matches the self-monitoring data, environmental data and dietary pictures corresponding to the service code of the object to be tested;

[0062] The self-monitoring data is brain wave data, and the brain wave data is represented by a brain wave image within a monitoring period;

[0063] The environmental data includes environmental noise, environmental humidity and environmental temperature;

[0064] The food pictures represent food pictures taken for lunch and food pictures taken for dinner.

[0065] In the specific implementation, by setting shooting parameters and adjusting the shooting angle and focal length, it is ensured that the proportion of effective pixels in the captured image reaches a certain threshold, thereby improving the accuracy of facial recognition. By automatically adjusting the shooting parameters, manual intervention is reduced, making the shooting process more automated and standardized, and improving efficiency. By utilizing the first gradient value and the second gradient value obtained by big data, the system can automatically adjust the shooting parameters according to different environments and conditions to enhance the adaptability and flexibility of the system. By continuously adjusting the shooting parameters until the requirements are met, it is ensured that users do not need to make manual adjustments, thereby improving the user experience. The captured image is bound to the service object code, and the label is set as the relative name, thereby ensuring the consistency and integrity of the data, facilitating subsequent data analysis and processing, and automatically adjusting the shooting parameters to reduce the number of repeated shooting due to unqualified shooting, saving time and resources.

[0066] The first analysis module divides the monitoring time into a first time period and a second time period according to the brain wave data;

[0067] The division logic of the first time period and the second time period includes:

[0068] The second value is set as the individual time step, and the third value is set as the time period, and the original brain wave signal is divided into sub-segments, the second value and the third value are obtained through big data, the second value represents the number of brain wave amplitudes contained in each sub-segment, and the third value represents the duration of each sub-segment, and the brain wave amplitudes in the sub-segments are obtained in time sequence, and the brain wave amplitudes are subjected to discrete wavelet transform to obtain the frequency domain expression of the amplitude, and the spectrum entropy value of the amplitude is calculated, and the average value of the spectrum entropy value of the amplitude is calculated and recorded as the first average value, and the standard spectrum entropy of the sleep period is obtained, and the first average value is compared with the standard spectrum entropy of the sleep period. When the first average value is greater than the standard spectrum entropy of the sleep period, the corresponding sub-segment is classified into the first time period, and when the first average value is less than or equal to the standard spectrum entropy of the sleep period, the corresponding sub-segment is classified into the second time period;

[0069] The first time point is set as the evening demarcation point, wherein the first time point is a generally recognized decomposition time point of evening and daytime, the sub-segments classified into the first time period and distributed before the first time point are set as the first time period, and the sub-segments classified into the second time period and distributed after the first time point are set as the second time period;

[0070] The first time period is represented as a time period of wakefulness, and the second time period is represented as a time period of sleep.

[0071] In specific implementation, by dividing the monitoring time into a first period of wakefulness and a second period of sleep, the monitoring time can be managed more finely, and different monitoring strategies can be provided for different time periods. The frequency domain expression of the EEG amplitude is obtained by discrete wavelet transform, and the spectrum entropy value is calculated. This method can more accurately capture the dynamic changes of brain waves, improve the accuracy of sleep staging and wakefulness state recognition, and compare the individual's brain wave characteristics with the standard spectrum entropy of the sleep period. It can provide each service object with a personalized monitoring strategy to meet the needs of different individuals. By using the second and third values ​​obtained by big data, the system can adapt to different brain wave signal characteristics and enhance the system's adaptability and generalization capabilities.

[0072] The first analysis module obtains brain wave data when the same first facial image appears in a first time period, recorded as first data, and counts the interval length of adjacent first data in the first time period, recorded as first duration, where the adjacent means adjacent in time order;

[0073] According to the first data and the first duration, an emotion impact model of the first facial image on the service object is constructed, and a corresponding relative visit decision is generated according to the emotion impact model. The construction logic of the emotion impact model includes:

[0074] Acquire first data, set a first segmentation step length, wherein the first segmentation step length is obtained through historical experience, segment the first data to obtain first sub-data, perform band-pass filtering on the first sub-data, extract the frequency characteristic quantity of the first sub-data after band-pass filtering, record it as the first characteristic quantity, obtain the frequency characteristic quantity of the standard α wave and the frequency characteristic quantity of the standard β wave, record it as the second characteristic quantity and the third characteristic quantity, set the third value as the similarity threshold, calculate the similarity between the first characteristic quantity and the second characteristic quantity through the cosine similarity formula, record it as the first similarity, and calculate the similarity between the first characteristic quantity and the third characteristic quantity, record it as the second similarity, and compare the second similarity with the first similarity;

[0075] When the first similarity is greater than or equal to the third value, the first sub-data is set to the α wave; when the second similarity is greater than or equal to the third value, the first sub-data is set to the β wave; otherwise, the first sub-data is deleted;

[0076] The number of α wave appearances and the number of β wave appearances are counted respectively, recorded as the third number and the fourth number, the product of the third number and the segmentation step length, and the product of the fourth number and the segmentation step length are calculated respectively, recorded as the first product and the second product, the ratio of the second product to the first product is calculated, recorded as the second ratio, the second ratio represents the degree of excitement, and the larger the second ratio is, the more excited the person is;

[0077] Taking the second ratio as the dependent variable and the first duration as the independent variable, the emotional impact model was constructed through mathematical tools;

[0078] Set an excitement interval, obtain an upper limit value of the first duration and a lower limit value of the first duration according to the excitement interval, and obtain a corresponding relative visit decision according to the upper limit value of the first duration and the lower limit value of the first duration, wherein the corresponding relative visit decision includes a number X i For relatives, the minimum visit interval is the lower limit of the first duration, and the maximum visit interval is the upper limit of the first duration.

[0079] The calculation expression of the emotion impact model is:

[0080]

[0081] in, Expressed as The second ratio corresponding to the relatives, Represents the first duration, The transformation rule expressed as the ratio of the first duration to the second is obtained by mathematical tools.

[0082] In specific implementation, by analyzing the alpha waves and beta waves in the brain wave data, the emotional state of the service recipients can be monitored. The alpha waves are usually associated with a relaxed state, while the beta waves are associated with an alert and excited state. Using the individual's brain wave data to construct an emotional impact model can more accurately reflect the emotional response of the service recipients to specific stimuli (such as relatives' visits). The relative visiting decisions generated by the emotional impact model can optimize the visiting plan to ensure that the service recipients receive emotional support when needed while avoiding excessive disturbances. Bandpass filtering and frequency feature extraction improve the accuracy of brain wave data analysis, making the judgment of emotional state more reliable. By calculating the degree of excitement (the second ratio) and using it as the dependent variable, combined with the first duration as the independent variable, personalized emotional support matching can be provided for each service recipient. Setting the visiting interval according to the excitement range can improve the effectiveness of relative visits and ensure that the service recipients receive emotional support at the best time.

[0083] The first analysis module determines the sleep quality level according to the brain wave data in the second time period, and the sleep quality level determination logic includes:

[0084] Obtain brain wave data in a second time period, set a second segmentation step, the second segmentation step is obtained through historical experience, the brain wave data in the second time period is segmented to obtain second sub-data, the α wave power and the β wave power in each second sub-data are obtained, the average value of the α wave power and the average value of the β wave power are calculated respectively, which are recorded as the second average value and the third average value respectively, the ratio of the third average value to the second average value is calculated, which is recorded as the third ratio, the fourth value and the fifth value are set as the threshold of the third ratio, the fourth value and the fifth value are obtained through historical experience, the third ratio is compared with the fourth value and the fifth value, when the third ratio is less than the fourth value, the sleep quality level is set to the first level, when the third ratio is greater than or equal to the fourth value and less than the fifth value, the sleep quality level is set to the second level, when the third ratio is greater than or equal to the fifth value, the sleep quality level is set to the third level, and the sleep quality represented by the first level, the second level and the third level gradually decreases;

[0085] A sleep quality impact model is constructed according to the sleep quality level, environmental noise, environmental humidity and environmental temperature, and an environmental control decision is generated according to the sleep quality impact model.

[0086] In specific implementation, by analyzing the power of alpha and beta waves in brain wave data, sleep quality can be evaluated more accurately, providing a scientific basis for subsequent environmental regulation. According to the thresholds set based on individual brain wave data and historical experience, personalized sleep management plans can be provided for each service recipient. Combined with the sleep quality level and environmental factors (noise, humidity, temperature), targeted environmental regulation decisions can be generated to optimize the sleep environment. More refined sleep management and environmental regulation can be achieved through scientific and technological means, thereby improving the service quality of nursing homes. The application of this technology has promoted the development of health monitoring technology, especially in the field of remote monitoring of nursing homes, providing more scientific and detailed monitoring for the elderly.

[0087] The construction logic of the sleep quality impact model includes:

[0088] Taking the sleep quality level as the dependent variable and the ambient noise, ambient humidity and ambient temperature as the independent variables, a multiple regression analysis was performed to obtain a sleep quality impact model. The calculation expression of the sleep quality impact model is:

[0089]

[0090] in, is the sleep quality level, For the independent variables, For the The coefficients corresponding to the independent variables;

[0091] The current sleep quality level is obtained according to the sleep quality impact model. When the current sleep quality level is not the first level, the third gradient value and the fourth gradient value are set as the change in ambient humidity and the change in ambient temperature respectively, and the ambient temperature interval and the ambient humidity interval are set. Within the ambient temperature interval and the ambient humidity interval, the ambient humidity and the ambient temperature are continuously regulated according to the third gradient value and the fourth gradient value, and the current sleep quality level after regulation is continuously obtained. When the current sleep quality level is the first level, the continuous regulation of the ambient humidity and the ambient temperature is stopped. Otherwise, a sound insulation device is added, and the sound insulation device includes a sound insulation board or a sound insulation green belt. The width and height of the sound insulation device are continuously changed, and the current sleep quality level is changed to the first level.

[0092] In the specific implementation, through multiple regression analysis, with sleep quality level as the dependent variable, environmental noise, environmental humidity and ambient temperature as independent variables, the constructed model can scientifically evaluate the extent to which sleep quality is affected by environmental factors. By calculating the coefficients corresponding to each independent variable, the specific impact of environmental noise, humidity and temperature on sleep quality is quantified, providing accurate data support for environmental control. According to the current sleep quality level obtained by the model, the system can adjust the ambient humidity and temperature in real time to optimize the sleeping environment and improve sleep quality. When environmental control cannot achieve ideal sleep quality, the system will add sound insulation devices, such as sound insulation boards or sound insulation green belts, to further improve the sleeping environment.

[0093] The second analysis module is connected to the hospital system, inputs the service code of the service object into the hospital system, and obtains the doctor's order information of the service object, wherein the doctor's order information is expressed as dietary advice, and the dietary advice is expressed as the name of food that cannot be eaten.

[0094] Identify food names according to the diet picture, compare the food names with the diet recommendations to obtain a first comparison result, and perform a first operation according to the first comparison result, wherein the first comparison result includes whether the food name meets the diet recommendations and whether the food name does not meet the diet recommendations, and the first operation includes not sending a food warning signal and sending a food warning signal;

[0095] When the first comparison result is that the food name does not conform to the dietary recommendations, the first operation is set to send a food warning signal; otherwise, the first operation is set to not send a food warning signal.

[0096] In specific implementation, personalized dietary management can be carried out according to the specific dietary recommendations of the service recipients to ensure that their diet complies with the doctor's orders or nutritional needs. By identifying foods that do not meet the dietary recommendations and sending early warning signals, the system helps prevent the service recipients from ingesting foods that may have adverse effects on their health. It provides real-time dietary monitoring and feedback, allowing caregivers to respond quickly and take necessary measures, such as adjusting diet plans or providing dietary guidance. Family members and caregivers can use the system to keep abreast of the service recipients' dietary status, thereby enhancing their confidence and satisfaction with the service recipients' dietary management. By preventing diet-related health problems, the system helps reduce medical costs caused by improper diet.

[0097] By collecting facial images and related facial images of the service object, the present invention can identify the service code of the object to be tested and match the corresponding monitoring data, which is helpful to quickly locate and respond to the needs of the elderly and improve monitoring efficiency. The monitoring time is divided into different time periods according to the brain wave data, and an emotional impact model is constructed, which is helpful to monitor the emotional state of the elderly and generate corresponding relative visit decisions according to the model, so as to provide more personalized care and support. The sleep quality impact model can be constructed according to the sleep quality level and environmental data, and environmental regulation decisions can be generated, which is helpful to improve the sleeping environment of the elderly and improve the sleep quality. The medical advice information of the service object is obtained, and the name of the diet is identified according to the diet picture, and compared with the diet suggestion to ensure that the diet of the elderly complies with the doctor's advice, which helps the elderly to better follow the doctor's advice and promote health. The smart health equipment can also realize the health record management of the elderly, understand the changing trend of the health status of the elderly, and comprehensively evaluate the health status of the elderly.

[0098] It should be understood by those skilled in the art that the embodiments of the present invention may be provided as methods, systems or computer program products. Therefore, the present invention may take the form of a complete hardware embodiment, a complete software embodiment or an embodiment combining software and hardware aspects. Moreover, the present invention may take the form of a computer program product implemented on one or more computer-usable storage media containing computer-usable program codes. Among them, the storage medium may be implemented by any type of volatile or non-volatile storage device or a combination thereof, such as static random access memory (Static Random Access Memory, referred to as SRAM), electrically erasable programmable read-only memory (Electrically Erasable Programmable Read-Only Memory, referred to as EEPROM), erasable programmable read-only memory (Erasable Programmable Read Only Memory, referred to as EPROM), programmable read-only memory (Programmable Red-Only Memory, referred to as PROM), read-only memory (Read-Only Memory, referred to as ROM), magnetic memory, flash memory, magnetic disk or optical disk. These computer program instructions may also be stored in a computer-readable memory capable of directing a computer or other programmable data processing device to operate in a specific manner, so that the instructions stored in the computer-readable memory produce an article of manufacture comprising an instruction device, which implements the process Figure 1 A process or multiple processes and / or boxes Figure 1 A function specified in one or more boxes.

[0099] It should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention rather than to limit it. Although the present invention has been described in detail with reference to the preferred embodiments, those skilled in the art should understand that the technical solutions of the present invention may be modified or replaced by equivalents without departing from the spirit and scope of the technical solutions of the present invention, which should all be included in the scope of the claims of the present invention.

Claims

1. A remote monitoring system for nursing homes based on image processing, characterized in that: It includes a collection module, a first analysis module and a second analysis module; The acquisition module records the service object's own facial image and related facial images, identifies the service code of the subject to be tested, matches the self-monitoring data, environmental data and dietary pictures corresponding to the service code of the subject to be tested, numbers the related facial images, and obtains a first facial image; The first analysis module divides the monitoring time into a first time period and a second time period according to the brain wave data, obtains the self-monitoring data when the same first facial image appears in the first time period, recorded as the first data, calculates the interval length of adjacent first data in the first time period, recorded as the first duration, constructs an emotional impact model of the first facial image on the service object according to the first data and the first duration, generates a corresponding relative visit decision according to the emotional impact model, determines the sleep quality level according to the self-monitoring data in the second time period, constructs a sleep quality impact model according to the sleep quality level and environmental data, and generates an environmental control decision according to the sleep quality impact model; The second analysis module obtains the doctor's advice information of the service object, identifies the name of the diet according to the diet picture, compares the diet name with the diet suggestion to obtain a first comparison result, and performs a first operation according to the first comparison result.

2. A remote monitoring system for nursing homes based on image processing as claimed in claim 1, characterized in that: The acquisition module records the service object's own facial image and related facial images, the related facial images are facial images of the service object's relatives, and identifies the service code of the object to be tested, the service code of the object to be tested is sensed by radio frequency identification through the service object wearing a bracelet, and the recording logic of the own facial image and related facial images includes: Starting a face recording device, setting shooting parameters, wherein the shooting parameters include a shooting angle and a shooting focal length, sending a recording start signal, shooting a face, obtaining a captured image, identifying a face contour in the captured image, counting the number of pixels within the face contour, recorded as a first number, counting the total number of pixels in the captured image, recorded as a second number, calculating a ratio of the first number to the second number, recorded as a first ratio, wherein the first ratio is expressed as a proportion of valid pixels in the captured image, setting a first value as a threshold of the first ratio, and comparing the first ratio with the first value; When the first ratio is greater than or equal to the first value, the captured image is bound to the service object code, and the label of the captured image is set to the relative name of the service object; When the first ratio is less than the first value, the first gradient value is set as the change in the shooting angle, the second gradient value is set as the change in the shooting focal length, the shooting parameters are continuously adjusted according to the first gradient value and the second gradient value, and the adjusted first ratio is continuously calculated until the adjusted first ratio is greater than or equal to the first value, then the continuous adjustment of the shooting parameters is stopped, the adjusted shooting image is bound to the service object code, and the label of the adjusted shooting image is set to the relative name of the service object; The continuous control is represented by continuous increase or continuous decrease, and the first gradient value and the second gradient value are obtained through big data; When the captured image is bound to the service object code and the label of the captured image is set, the face entry device is closed; The related facial images are numbered to obtain a first facial image, and the number of the related facial image is represented by X i , where i is a natural number, X i Represents the relatives of the service object whose face image is entered.

3. A remote monitoring system for nursing homes based on image processing as claimed in claim 2, characterized in that: The acquisition module retrieves the service database, inputs the service object code into the service database, and matches the self-monitoring data, environmental data and dietary pictures corresponding to the service code of the object to be tested; The self-monitoring data is brain wave data, and the brain wave data is represented by a brain wave image within a monitoring period; The environmental data includes environmental noise, environmental humidity and environmental temperature; The food pictures represent food pictures taken for lunch and food pictures taken for dinner.

4. The remote monitoring system for nursing homes based on image processing as claimed in claim 1, characterized in that: The first analysis module divides the monitoring time into a first time period and a second time period according to the brain wave data; The division logic of the first time period and the second time period includes: The second value is set as the individual time step, and the third value is set as the time period, and the original brain wave signal is divided into sub-segments, the second value and the third value are obtained through big data, the second value represents the number of brain wave amplitudes contained in each sub-segment, and the third value represents the duration of each sub-segment, and the brain wave amplitudes in the sub-segments are obtained in time sequence, and the brain wave amplitudes are subjected to discrete wavelet transform to obtain the frequency domain expression of the amplitude, and the spectrum entropy value of the amplitude is calculated, and the average value of the spectrum entropy value of the amplitude is calculated and recorded as the first average value, and the standard spectrum entropy of the sleep period is obtained, and the first average value is compared with the standard spectrum entropy of the sleep period. When the first average value is greater than the standard spectrum entropy of the sleep period, the corresponding sub-segment is classified into the first time period, and when the first average value is less than or equal to the standard spectrum entropy of the sleep period, the corresponding sub-segment is classified into the second time period; The first time point is set as the evening demarcation point, wherein the first time point is a generally recognized decomposition time point of evening and daytime, the sub-segments classified into the first time period and distributed before the first time point are set as the first time period, and the sub-segments classified into the second time period and distributed after the first time point are set as the second time period; The first time period is represented as a time period of wakefulness, and the second time period is represented as a time period of sleep.

5. The remote monitoring system for nursing homes based on image processing as claimed in claim 3, characterized in that: The first analysis module obtains brain wave data when the same first facial image appears in a first time period, recorded as first data, and counts the interval length of adjacent first data in the first time period, recorded as first duration, where the adjacent means adjacent in time order; According to the first data and the first duration, an emotion impact model of the first facial image on the service object is constructed, and a corresponding relative visit decision is generated according to the emotion impact model. The construction logic of the emotion impact model includes: Acquire first data, set a first segmentation step length, wherein the first segmentation step length is obtained through historical experience, segment the first data to obtain first sub-data, perform band-pass filtering on the first sub-data, extract the frequency characteristic quantity of the first sub-data after band-pass filtering, record it as the first characteristic quantity, obtain the frequency characteristic quantity of the standard α wave and the frequency characteristic quantity of the standard β wave, record it as the second characteristic quantity and the third characteristic quantity, set the third value as the similarity threshold, calculate the similarity between the first characteristic quantity and the second characteristic quantity through the cosine similarity formula, record it as the first similarity, and calculate the similarity between the first characteristic quantity and the third characteristic quantity, record it as the second similarity, and compare the second similarity with the first similarity; When the first similarity is greater than or equal to the third value, the first sub-data is set to the α wave; when the second similarity is greater than or equal to the third value, the first sub-data is set to the β wave; otherwise, the first sub-data is deleted; The number of α wave appearances and the number of β wave appearances are counted respectively, recorded as the third number and the fourth number, the product of the third number and the segmentation step length, and the product of the fourth number and the segmentation step length are calculated respectively, recorded as the first product and the second product, the ratio of the second product to the first product is calculated, recorded as the second ratio, the second ratio represents the degree of excitement, and the larger the second ratio is, the more excited the person is; Taking the second ratio as the dependent variable and the first duration as the independent variable, the emotional impact model was constructed through mathematical tools; Set an excitement interval, obtain an upper limit value of the first duration and a lower limit value of the first duration according to the excitement interval, and obtain a corresponding relative visit decision according to the upper limit value of the first duration and the lower limit value of the first duration, wherein the corresponding relative visit decision includes a number X i For relatives, the minimum visit interval is the lower limit of the first duration, and the maximum visit interval is the upper limit of the first duration.

6. A remote monitoring system for nursing homes based on image processing as claimed in claim 5, characterized in that: The calculation expression of the emotion impact model is: ； in, Expressed as The second ratio corresponding to the relatives, Represents the first duration, The transformation rule expressed as the ratio of the first duration to the second is obtained by mathematical tools.

7. The remote monitoring system for nursing homes based on image processing as claimed in claim 5, characterized in that: The first analysis module determines the sleep quality level according to the brain wave data in the second time period, and the sleep quality level determination logic includes: Obtain brain wave data in a second time period, set a second segmentation step, the second segmentation step is obtained through historical experience, the brain wave data in the second time period is segmented to obtain second sub-data, the α wave power and the β wave power in each second sub-data are obtained, the average value of the α wave power and the average value of the β wave power are calculated respectively, which are recorded as the second average value and the third average value respectively, the ratio of the third average value to the second average value is calculated, which is recorded as the third ratio, the fourth value and the fifth value are set as the threshold of the third ratio, the fourth value and the fifth value are obtained through historical experience, the third ratio is compared with the fourth value and the fifth value, when the third ratio is less than the fourth value, the sleep quality level is set to the first level, when the third ratio is greater than or equal to the fourth value and less than the fifth value, the sleep quality level is set to the second level, when the third ratio is greater than or equal to the fifth value, the sleep quality level is set to the third level, and the sleep quality represented by the first level, the second level and the third level gradually decreases; A sleep quality impact model is constructed according to the sleep quality level, environmental noise, environmental humidity and environmental temperature, and an environmental control decision is generated according to the sleep quality impact model.

8. The remote monitoring system for nursing homes based on image processing as claimed in claim 7, characterized in that: The construction logic of the sleep quality impact model includes: Taking the sleep quality level as the dependent variable and the ambient noise, ambient humidity and ambient temperature as the independent variables, a multiple regression analysis was performed to obtain a sleep quality impact model. The calculation expression of the sleep quality impact model is: ； in, is the sleep quality level, For the independent variables, For the The coefficients corresponding to the independent variables; The current sleep quality level is obtained according to the sleep quality impact model. When the current sleep quality level is not the first level, the third gradient value and the fourth gradient value are set as the change in ambient humidity and the change in ambient temperature respectively, and the ambient temperature interval and the ambient humidity interval are set. Within the ambient temperature interval and the ambient humidity interval, the ambient humidity and the ambient temperature are continuously regulated according to the third gradient value and the fourth gradient value, and the current sleep quality level after regulation is continuously obtained. When the current sleep quality level is the first level, the continuous regulation of the ambient humidity and the ambient temperature is stopped. Otherwise, a sound insulation device is added, and the sound insulation device includes a sound insulation board or a sound insulation green belt. The width and height of the sound insulation device are continuously changed, and the current sleep quality level is changed to the first level.

9. The remote monitoring system for nursing homes based on image processing as claimed in claim 1, characterized in that: The second analysis module is connected to the hospital system, inputs the service code of the service object into the hospital system, and obtains the doctor's order information of the service object, wherein the doctor's order information is expressed as dietary advice, and the dietary advice is expressed as the name of food that cannot be eaten.

10. The remote monitoring system for nursing homes based on image processing as claimed in claim 9, characterized in that: Identify food names according to the diet picture, compare the food names with the diet recommendations to obtain a first comparison result, and perform a first operation according to the first comparison result, wherein the first comparison result includes whether the food name meets the diet recommendations and whether the food name does not meet the diet recommendations, and the first operation includes not sending a food warning signal and sending a food warning signal; When the first comparison result is that the food name does not conform to the dietary recommendations, the first operation is set to send a food warning signal; otherwise, the first operation is set to not send a food warning signal.

Images