Elderly care service management method and system

Abstract

The present invention discloses a method and system for managing elderly care services. The method comprises the following steps: collecting an initial color image and an initial infrared image of elderly clothes just worn, and a secondary color image and a secondary infrared image of the elderly clothes on the nth day; performing decolorization processing on the initial color image and the secondary color image, respectively, to obtain an initial colorless image and a secondary colorless image; obtaining the hygiene of the elderly clothes and an average hygiene value corresponding to a comprehensive hygiene situation according to the initial cleanliness image and the secondary cleanliness image; the technical scheme collects different environmental information and the overall hygiene partitions of the elderly clothes and the average hygiene values ​​corresponding to the cleanliness areas under the environmental information, and uses the situation information, the area of ​​the cleanliness area, and the average hygiene value as a training set to train a TCN planning network, calculate the cleanliness of the elderly clothes, and when the cleanliness is lower than a minimum cleanliness value, remind the staff to wash the elderly clothes.

Description

Technical Field

[0001] The present invention relates to the technical field of elderly care services, and in particular to an elderly care service management method and system. Background Art

[0002] As the population grows, aging becomes more and more serious, which has become a prominent social problem. The number of elderly people in the total population is increasing year by year. Further optimization of the elderly care service industry is imperative and has become a common concern of the public.

[0003] At present, home-based care or nursing home-based care is widely adopted. Home-based care allows the elderly to live with their families. In many cases, family members are working. In many cases, the elderly need to be alone and have high self-care ability. Nursing home-based care can be provided by specialized nursing staff and medical personnel, and a variety of services can be provided to ensure the elderly's food, clothing, housing and transportation, and the family members can also be relatively relieved. Usually, if the economy allows, they will choose nursing home-based care.

[0004] However, in the actual nursing home management work, when cleaning the clothes, children cannot pay real-time attention to the elderly's food, clothing, housing and transportation in the nursing home, nor do they report the elderly's situation in real time. Generally, they will learn about the elderly's recent situation after a period of time. Based on this situation, many nursing homes will be lax in management and will not regularly wash the elderly's clothes. In many cases, the clothes are already very dirty. If the elderly wear these clothes for a long time, it is easy for bacteria to breed, viruses to be infected, and affect their health. Summary of the invention

[0005] The present invention provides a method and system for managing elderly care services. The technical solution collects different situation information and the overall hygiene zones and the average hygiene values ​​corresponding to the cleanliness zones of the elderly's clothes under the situation information. The situation information, the cleanliness zone area and the average hygiene value are used as a training set to train the TCN planning network, and then the cleanliness of the elderly's clothes is calculated. When the cleanliness is lower than the minimum cleanliness value, a prompt is turned on to remind the staff to clean the elderly's clothes in time, thereby solving the problems existing in the management of existing nursing homes, such as the inability to regularly clean the elderly's clothes and affecting the health of the elderly.

[0006] The present invention is achieved through the following technical solutions:

[0007] A method for managing elderly care services comprises the following steps:

[0008] A001: Collect the initial color image and initial infrared image of the old man's clothes when they are just put on, and the secondary color image and secondary infrared image of the old man's clothes on the nth day;

[0009] A002: Decoloring the initial color image and the secondary color image to obtain an initial colorless image and a secondary colorless image, and sending the initial colorless image and the initial infrared image to the hygiene service supervision network area to obtain an initial cleanliness image of the elderly's clothes;

[0010] A003: Send the secondary colorless image and the secondary infrared image to the hygiene service supervision network area to obtain the secondary cleanliness image of the elderly's clothes on the nth day;

[0011] A004: Obtaining the average hygiene value corresponding to the hygiene degree of the elderly's clothes and the comprehensive hygiene condition according to the initial cleanliness image and the secondary cleanliness image;

[0012] A005: Obtain information about the clothes worn by the elderly from the time they were first worn to the nth day, establish a training set for each piece of information and the overall cleanliness area and hygiene average value of the clothes in the corresponding situation, obtain multiple training sets, and use the training sets to perform TCN planning network training;

[0013] A006: Obtain real-time information on the clothes worn by the elderly in the next time period, and send it to the TCN planning network for practice, so as to obtain the regional area estimation value and the average hygiene estimation value of the overall cleanliness of the elderly's clothes in the next time period, and estimate the cleanliness of the elderly's clothes through the regional area estimation value and the average hygiene estimation value.

[0014] Optionally, in step A002, the initial colorless image and the initial infrared image are sent to the hygiene service supervision network area to obtain the initial cleanliness image of the elderly's clothes;

[0015] Among them, the hygiene service supervision network area includes two supervision network partitions, and the colorless image and the initial infrared image are respectively transmitted to a supervision network partition. The infrared image corresponds to the re-partitioned part in the supervision network partition, and the re-partitioned part is an encoding device-decoding device. The re-partitioned part of the supervision network partition corresponding to the colorless image is an encoding device, and the encoding devices in the two supervision network partitions have the same structure.

[0016] Optionally, the calculation results of the two supervisory network partitions are integrated with the output results of the supervisory network partitions using a CAM algorithm, and then processed using a decoding device to obtain a cleanliness image of the same size as the initial image.

[0017] Optionally, in step A002, the initial colorless image and the initial infrared image are sent to the hygiene service supervision network area to obtain the initial cleanliness image of the elderly's clothes at the beginning, and the specific operations are:

[0018] A0021: sending the initial infrared image to a supervision network sub-area of ​​the hygiene service supervision network area, using an encoding device to obtain a comprehensive processing result image of the initial infrared image, and using a decoding device to sample and process the comprehensive processing result image to obtain a characteristic colorless image;

[0019] A0022: taking the resolution value of each frame of the characteristic colorless image as the characteristic degree of each frame of the image, the characteristic degree being the probability that each frame of the image belongs to a hygiene characteristic point;

[0020] A0023: sending the initial colorless image to the sub-area of ​​the hygiene service supervision network area, and then obtaining a plurality of comprehensive processing result graphs through an encoding device, and processing the comprehensive processing result graphs and the characteristic colorless image using a CAM algorithm to obtain a composition characteristic graph;

[0021] A0024: The composition characteristic diagram is collected and processed by a decoding device, and the hygiene service supervision network area obtains an initial cleanliness image of the same size as the composition characteristic diagram, and the initial cleanliness image when the elderly person's clothes are just put on is segmented to obtain the cleanliness area when the elderly person's clothes are just put on, thereby obtaining the cleanliness area corresponding to the cleanliness area, and according to the resolution value of each frame of the image corresponding to the cleanliness area, the colorless average value corresponding to the cleanliness area when the elderly person's clothes are just put on is obtained.

[0022] Optionally, step A0024 also includes: sending the secondary color image and the secondary infrared image to the hygiene service supervision network area, obtaining the cleanliness area of ​​the elderly's clothes on the nth day of wearing, thereby obtaining the secondary cleanliness area corresponding to the cleanliness area when the elderly's clothes are worn on the nth day, and according to the resolution value of each frame of the image corresponding to the cleanliness area, obtaining the secondary colorless average value corresponding to the cleanliness area of ​​the elderly's clothes on the nth day of wearing.

[0023] Optionally, the overall cleanliness area of ​​the elderly's clothes and the average sanitation degree of the overall cleanliness area are obtained according to the cleanliness areas when the elderly's clothes are just worn and when the elderly's clothes are worn for the nth day, wherein the overall cleanliness area is calculated by adding the cleanliness area corresponding to the cleanliness area when the elderly's clothes are just worn and the secondary cleanliness area corresponding to the cleanliness area when the elderly's clothes are worn for the nth day to obtain the overall cleanliness area;

[0024] The method for obtaining the average sanitation value of the overall cleanliness area is: calculate the colorless average value corresponding to the cleanliness area when the elderly person's clothes are just worn and the secondary colorless average value corresponding to the cleanliness area when the elderly person's clothes are worn for the nth day, and use the average value as the sanitation average value of the entire cleanliness area.

[0025] Optionally, in step A005, the situation information includes dust quantity, humidity value, and temperature value.

[0026] Optionally, in step A005, a training set is established for each situation information and the overall cleanliness area and hygiene average value of the elderly's clothes in the corresponding situation to obtain multiple training sets, and the training sets are used to perform TCN planning network training, wherein the specific operation steps of performing TCN planning network training are:

[0027] A0051: Send the information about the elderly's clothes from the time they are first worn to the nth day to the TCN planning network, and obtain the average hygiene values ​​corresponding to the overall hygiene zones and overall cleanliness zones of the elderly's clothes in the next time period;

[0028] A0052: The convolution structure shape and size of the TCN planning network is: [2,4]. The encoding device is used to obtain the associated characteristic data within a period of time, and the overall hygiene partition of the elderly's clothes in the next time period and the average hygiene value corresponding to the overall cleanliness area are obtained through the associated network.

[0029] Optionally, in step A006, the real-time situation information of the clothes worn by the elderly in the next time period is obtained, and input into the TCN planning network for practice, so as to obtain the regional area estimation value and the average hygiene estimation value of the overall cleanliness of the elderly's clothes in the next time period;

[0030] According to the estimated value of the regional area and the estimated value of the average hygiene level, the cleanliness of the elderly's clothes in the next time period is calculated, that is, B=C+F, wherein B is the cleanliness level, C is the estimated value of the regional area, and F is the estimated value of the average hygiene level. When the cleanliness level is lower than the minimum cleanliness level, a prompt is turned on to remind the staff to wash the elderly's clothes in time.

[0031] The technical solution also provides an elderly care service management system in which the elderly care service management method described above is applied to the elderly care field.

[0032] Compared with the prior art, the present invention has the following advantages and beneficial effects:

[0033] 1) This technical solution collects different situation information and the overall hygiene zones and the average hygiene values ​​corresponding to the cleanliness zones of the elderly's clothes under the situation information. The situation information, cleanliness zone area, and average hygiene values ​​are used as training sets to train the TCN planning network, and then the cleanliness of the elderly's clothes is calculated. When the cleanliness is lower than the minimum cleanliness value, a prompt is turned on to remind the staff to clean the elderly's clothes in time.

[0034] 2) This technical solution collects infrared images and uses the radiation energy of the infrared images to analyze the corresponding images, ensuring that the area of ​​the estimated cleanliness area and the hygiene value corresponding to each frame of the image are more accurate, thereby ensuring that the results of calculating the cleanliness are more reliable. BRIEF DESCRIPTION OF THE DRAWINGS

[0035] Figure 1 A flow chart showing the structure of Example 1 of the present invention. DETAILED DESCRIPTION

[0036] In order to make the purpose of the present invention - technical solutions and advantages more clearly understood, the present invention is further described in detail below in conjunction with embodiments. The illustrative embodiments of the present invention and their descriptions are only used to explain the present invention and are not intended to limit the present invention.

[0037] Embodiment 1:

[0038] like Figure 1 As shown, a method for managing elderly care services comprises the following steps:

[0039] A001: Collect the initial color image and initial infrared image of the old man's clothes when they are just put on, and the secondary color image and secondary infrared image of the old man's clothes on the nth day;

[0040] A002: Decoloring the initial color image and the secondary color image to obtain an initial colorless image and a secondary colorless image, and sending the initial colorless image and the initial infrared image to the hygiene service supervision network area to obtain an initial cleanliness image of the elderly's clothes;

[0041] A003: Send the secondary colorless image and the secondary infrared image to the hygiene service supervision network area to obtain the secondary cleanliness image of the elderly's clothes on the nth day;

[0042] A004: Obtaining the average hygiene value corresponding to the hygiene degree of the elderly's clothes and the comprehensive hygiene condition according to the initial cleanliness image and the secondary cleanliness image;

[0043] Specifically, as the aging population becomes more and more serious, some families will send their elderly to nursing homes for care, which is conducive to improving various aspects of elderly care services, such as medical care, nursing, and management. However, due to the management of nursing homes, they cannot provide effective services to the elderly on a regular basis. For example, since there is no need to report the elderly's situation to their children in real time, the children cannot track the elderly's situation in the nursing home in real time. Therefore, nursing homes cannot supervise the elderly's food, clothing, housing and transportation in service management. For example, the elderly's clothes cannot be washed regularly. Some elderly people have poor self-care ability due to their age. Combined with the fact that nursing homes cannot regularly wash clothes and provide care, it is easy to affect the elderly's physical health.

[0044] Therefore, in this embodiment, the initial color image and the initial infrared image when the elderly just put on the clothes are collected, as well as the secondary color image and the secondary infrared image of the elderly on the nth day of wearing the clothes, and the initial color image and the secondary color image are processed to obtain the initial colorless image and the secondary colorless image. In this embodiment, the initial color image, the initial infrared image, the secondary color image and the secondary infrared image respectively obtained when the elderly just put on the clothes and the elderly on the nth day of wearing the clothes are obtained by shooting with a surveillance camera with an infrared lens.

[0045] The initial color image and the secondary color image are subjected to colorless or grayscale processing respectively to obtain an initial colorless image and a secondary colorless image. The specific method of colorless processing is: converting the color image into an HSV image by color region conversion, retaining the H color channel image corresponding to the HSV image, and setting the values ​​of the remaining two channels to 0, thereby obtaining a colorless image.

[0046] A005: Obtain information about the clothes worn by the elderly from the time they were first worn to the nth day, establish a training set for each piece of information and the overall cleanliness area and hygiene average value of the clothes in the corresponding situation, obtain multiple training sets, and use the training sets to perform TCN planning network training;

[0047] Specifically, a continuous time period is used, such as when the elderly wear clothes for the first time, from the first day to the nth day when the elderly wear clothes, to collect statistics on the environmental information of the elderly, and the statistical situation information, environmental information, overall cleanliness, and hygiene mean values ​​together constitute a training set sample. Based on a continuous time period, a training set corresponding to the image corresponding information collected for the elderly's clothes in multiple time periods is obtained, and multiple training sets are used to perform TCN planning network training.

[0048] A006: Obtain real-time information on the clothes worn by the elderly in the next time period, and send it to the TCN planning network for practice, so as to obtain the regional area estimation value and the average hygiene estimation value of the overall cleanliness of the elderly's clothes in the next time period, and estimate the cleanliness of the elderly's clothes through the regional area estimation value and the average hygiene estimation value.

[0049] Specifically, after obtaining the real-time situation information of the elderly's clothes in the next time period, the real-time situation information is input into the TCN planning network for training, and the regional area estimation value and the average hygiene estimation value of the overall cleanliness of the elderly's clothes in the next time period are obtained. The cleanliness of the elderly's clothes in the specific next time period is calculated by the regional area estimation value and the average hygiene estimation value. At the same time, the minimum cleanliness value is set. If the cleanliness value of the elderly's clothes is lower than the minimum cleanliness value, an early warning is activated to remind the nursing staff that timely care is needed.

[0050] Among them, the TCN planning network is a special convolutional neural network, specifically a time series convolutional network, which has a longer memory and can use the input training information to further deepen the training. It can predict the overall cleanliness and hygiene of the clothes worn by the elderly in any time period or any time period based on the real-time environmental information or situation information of the clothes worn by the elderly in the next time period, and issue prompts based on the prediction results to assist the nursing home in good management and reminders to ensure that the elderly's clothes are taken care of in time. At the same time, the TCN planning network has significant effects in the planning of time series data and has a high prediction accuracy.

[0051] By performing TCN planning network exercises on multiple training sets, the time series is classified according to the time period, and information is extracted from historical data. Based on the potential correlation of the extracted information in the time period, the hygiene of the elderly's clothes in the next historical period in the future can be accurately predicted.

[0052] Optionally, in step A002, the initial colorless image and the initial infrared image are sent to the hygiene service supervision network area to obtain the initial cleanliness image of the elderly's clothes;

[0053] Among them, the hygiene service supervision network area includes two supervision network partitions, and the colorless image and the initial infrared image are respectively transmitted to a supervision network partition. The infrared image corresponds to the re-partitioned part in the supervision network partition, and the re-partitioned part is an encoding device-decoding device. The re-partitioned part of the supervision network partition corresponding to the colorless image is an encoding device, and the encoding devices in the two supervision network partitions have the same structure.

[0054] The hygiene service supervision network area is specifically a model, that is, a neural network model. The neural network model adjusts the connection relationship between a large number of internal nodes through the complexity of the system, and processes the input information colorless image and initial infrared image in the supervision network partition, that is, data training is performed until convergence to obtain the hygiene service supervision network area, so that after processing, the initial cleanliness image of the clothes worn by the elderly is obtained, and the information is more accurate and more effective.

[0055] Optionally, the calculation results of the two supervisory network partitions are integrated with the output results of the supervisory network partitions using a CAM algorithm, and then processed using a decoding device to obtain a cleanliness image of the same size as the initial color image.

[0056] Specifically, the output results of the two supervision network partitions are composed of characteristic graphs by using the CAM algorithm, and then the composed characteristic graphs are comprehensively processed by a decoding device to obtain a cleanliness image of the same size as the initial color image. The loss function of the TCN planning network is an entropy loss function and a mean square error function.

[0057] Optionally, in step A002, the initial colorless image and the initial infrared image are sent to the hygiene service supervision network area to obtain the initial cleanliness image of the elderly's clothes at the beginning, and the specific operations are:

[0058] A0021: sending the initial infrared image to a supervision network sub-area of ​​the hygiene service supervision network area, using an encoding device to obtain a comprehensive processing result image of the initial infrared image, and using a decoding device to sample and process the comprehensive processing result image to obtain a characteristic colorless image;

[0059] A0022: taking the resolution value of each frame of the characteristic colorless image as the characteristic degree of each frame of the image, the characteristic degree being the probability that each frame of the image belongs to a hygiene characteristic point;

[0060] A0023: sending the initial colorless image to the sub-area of ​​the hygiene service supervision network area, and then obtaining a plurality of comprehensive processing result graphs through an encoding device, and processing the comprehensive processing result graphs and the characteristic colorless image using a CAM algorithm to obtain a composition characteristic graph;

[0061] A0024: The composition characteristic diagram is collected and processed by a decoding device, and the hygiene service supervision network area obtains an initial cleanliness image of the same size as the composition characteristic diagram, and the initial cleanliness image when the elderly person's clothes are just put on is segmented to obtain the cleanliness area when the elderly person's clothes are just put on, thereby obtaining the cleanliness area corresponding to the cleanliness area, and according to the resolution value of each frame of the image corresponding to the cleanliness area, the colorless average value corresponding to the cleanliness area when the elderly person's clothes are just put on is obtained.

[0062] Specifically, the initial infrared image includes some color texture information, which can reflect the magnitude of infrared energy radiation. Therefore, the energy generated by infrared radiation can accurately reflect the characteristic difference points between the pixels in the initial colorless image.

[0063] The resolution value of each frame in the characteristic colorless image is combined to obtain the characteristic degree of each frame, and the proportion of each frame is obtained according to the characteristic degree. The specific proportion is obtained by comprehensively calculating the number of pixels in the image, the proportion corresponding to each feature map, and the proportion coefficient of each pixel.

[0064] Optionally, step A0024 also includes: sending the secondary color image and the secondary infrared image to the hygiene service supervision network area, obtaining the cleanliness area of ​​the elderly's clothes on the nth day of wearing, thereby obtaining the secondary cleanliness area corresponding to the cleanliness area when the elderly's clothes are worn on the nth day, and according to the resolution value of each frame of the image corresponding to the cleanliness area, obtaining the secondary colorless average value corresponding to the cleanliness area of ​​the elderly's clothes on the nth day of wearing.

[0065] Optionally, the overall cleanliness area of ​​the elderly's clothes and the average sanitation degree of the overall cleanliness area are obtained according to the cleanliness areas when the elderly's clothes are just worn and when the elderly's clothes are worn for the nth day, wherein the overall cleanliness area is calculated by adding the cleanliness area corresponding to the cleanliness area when the elderly's clothes are just worn and the secondary cleanliness area corresponding to the cleanliness area when the elderly's clothes are worn for the nth day to obtain the overall cleanliness area;

[0066] The method for obtaining the average sanitation value of the overall cleanliness area is: calculate the colorless average value corresponding to the cleanliness area when the elderly person's clothes are just worn and the secondary colorless average value corresponding to the cleanliness area when the elderly person's clothes are worn for the nth day, and use the average value as the sanitation average value of the entire cleanliness area.

[0067] Optionally, in step A005, the situation information includes dust volume, humidity value, and temperature value. The situation information, such as dust volume, such as the road section passed, will also affect the cleanliness of the clothes worn; humidity value, such as rain or winter, will have a certain impact on the cleanliness of the clothes. Temperature value, such as summer, the temperature is too high, it is easy to sweat, and it is also necessary to regularly or even increase the frequency of care for the elderly's clothes.

[0068] Among them, data such as dust quantity, humidity value, temperature value, etc. can be obtained using existing equipment.

[0069] Optionally, in step A005, a training set is established for each situation information and the overall cleanliness area and hygiene average value of the elderly's clothes in the corresponding situation to obtain multiple training sets, and the training sets are used to perform TCN planning network training, wherein the specific operation steps of performing TCN planning network training are:

[0070] A0051: Send the information about the elderly's clothes from the time they are first worn to the nth day to the TCN planning network, and obtain the average hygiene values ​​corresponding to the overall hygiene zones and overall cleanliness zones of the elderly's clothes in the next time period;

[0071] A0052: The convolution structure shape and size of the TCN planning network is: [2,4]. The encoding device is used to obtain the associated characteristic data within a period of time, and the overall hygiene partition of the elderly's clothes in the next time period and the average hygiene value corresponding to the overall cleanliness area are obtained through the associated network.

[0072] Optionally, in step A006, the real-time situation information of the clothes worn by the elderly in the next time period is obtained, and input into the TCN planning network for practice, so as to obtain the regional area estimation value and the average hygiene estimation value of the overall cleanliness of the elderly's clothes in the next time period;

[0073] According to the estimated value of the regional area and the estimated value of the average hygiene level, the cleanliness of the elderly's clothes in the next time period is calculated, that is, B=C+F, wherein B is the cleanliness level, C is the estimated value of the regional area, and F is the estimated value of the average hygiene level. When the cleanliness level is lower than the minimum cleanliness level, a prompt is turned on to remind the staff to wash the elderly's clothes in time.

[0074] Embodiment 2:

[0075] This embodiment provides a pension service management system, which mainly applies the pension service method described in Embodiment 1 to the pension service management system.

[0076] The elderly care service management system collects information about different environmental conditions when the elderly just put on a piece of clothing, as well as the overall hygiene zones of the elderly's clothes and the average hygiene values ​​corresponding to the cleanliness zones under the environmental conditions. The situation information, cleanliness zone area, and hygiene average value are used as training sets to train the TCN planning network, and then the cleanliness of the elderly's clothes is calculated. When the cleanliness is lower than the minimum cleanliness value, a prompt is turned on to remind the staff to clean the elderly's clothes in time to avoid the failure to achieve the purpose of regular management of elderly services, which in turn affects the elderly's physical health.

[0077] The specific implementation methods described above further illustrate the purpose of the present invention - technical solutions and beneficial effects. It should be understood that the above description is only a specific implementation method of the present invention and is not intended to limit the scope of protection of the present invention. Any modifications - equivalent replacements - improvements, etc. made within the spirit and principles of the present invention should be included in the scope of protection of the present invention.

Claims

1. A method for managing elderly care services, characterized in that: The following steps are included: A001: Collect the initial color image and initial infrared image of the old man's clothes when they are just put on, and the secondary color image and secondary infrared image of the old man's clothes on the nth day; A002: Decoloring the initial color image and the secondary color image to obtain an initial colorless image and a secondary colorless image, and sending the initial colorless image and the initial infrared image to the hygiene service supervision network area to obtain an initial cleanliness image of the elderly's clothes; A003: Send the secondary colorless image and the secondary infrared image to the hygiene service supervision network area to obtain the secondary cleanliness image of the elderly's clothes on the nth day; A004: Obtaining the average hygiene value corresponding to the hygiene degree of the elderly's clothes and the comprehensive hygiene condition according to the initial cleanliness image and the secondary cleanliness image; A005: Obtain information about the clothes worn by the elderly from the time they were first worn to the nth day, establish a training set for each piece of information and the overall cleanliness area and hygiene average value of the clothes in the corresponding situation, obtain multiple training sets, and use the training sets to perform TCN planning network training; A006: Obtain real-time information on the clothes worn by the elderly in the next time period, and send it to the TCN planning network for practice, so as to obtain the regional area estimation value and the average hygiene estimation value of the overall cleanliness of the elderly's clothes in the next time period, and estimate the cleanliness of the elderly's clothes through the regional area estimation value and the average hygiene estimation value.

2. A method for managing elderly care services according to claim 1, characterized in that: In step A002, the initial colorless image and the initial infrared image are sent to the hygiene service supervision network area to obtain the initial cleanliness image of the elderly's clothes; Among them, the hygiene service supervision network area includes two supervision network partitions, and the colorless image and the initial infrared image are respectively transmitted to a supervision network partition. The infrared image corresponds to the re-partitioned part in the supervision network partition, and the re-partitioned part is an encoding device-decoding device. The re-partitioned part of the supervision network partition corresponding to the colorless image is an encoding device, and the encoding devices in the two supervision network partitions have the same structure.

3. A method for managing elderly care services according to claim 2, characterized in that: The calculation results of the two supervisory network partitions are integrated with the output results of the supervisory network partitions using the CAM algorithm, and then processed using a decoding device to obtain a cleanliness image of the same size as the initial image.

4. A method for managing elderly care services according to claim 3, characterized in that: In step A002, the initial colorless image and the initial infrared image are sent to the hygiene service supervision network area to obtain the initial cleanliness image of the elderly's clothes. The specific operations are: A0021: sending the initial infrared image to a supervision network sub-area of ​​the hygiene service supervision network area, using an encoding device to obtain a comprehensive processing result image of the initial infrared image, and using a decoding device to sample and process the comprehensive processing result image to obtain a characteristic colorless image; A0022: taking the resolution value of each frame of the characteristic colorless image as the characteristic degree of each frame of the image, the characteristic degree being the probability that each frame of the image belongs to a hygiene characteristic point; A0023: sending the initial colorless image to the sub-area of ​​the hygiene service supervision network area, and then obtaining a plurality of comprehensive processing result graphs through an encoding device, and processing the comprehensive processing result graph and the characteristic colorless image using a CAM algorithm to obtain a composition characteristic graph; A0024: The composition characteristic diagram is collected and processed by a decoding device, and the hygiene service supervision network area obtains an initial cleanliness image of the same size as the composition characteristic diagram, and the initial cleanliness image when the elderly person's clothes are just put on is segmented to obtain the cleanliness area when the elderly person's clothes are just put on, thereby obtaining the cleanliness area corresponding to the cleanliness area, and according to the resolution value of each frame of the image corresponding to the cleanliness area, the colorless average value corresponding to the cleanliness area when the elderly person's clothes are just put on is obtained.

5. A method for managing elderly care services according to claim 4, characterized in that: Step A0024 also includes: sending the secondary color image and the secondary infrared image to the hygiene service supervision network area, obtaining the cleanliness area of ​​the elderly's clothes on the nth day of wearing, thereby obtaining the secondary cleanliness area corresponding to the cleanliness area when the elderly's clothes are worn on the nth day, and according to the resolution value of each frame of the image corresponding to the cleanliness area, obtaining the secondary colorless average value corresponding to the cleanliness area of ​​the elderly's clothes on the nth day of wearing.

6. A method for managing elderly care services according to claim 5, characterized in that: According to the cleanliness area of ​​the elderly's clothes when they are just worn and the clothes on the nth day of wearing, the overall cleanliness area of ​​the elderly's clothes and the average hygiene value of the overall cleanliness area are obtained, wherein the overall cleanliness area is calculated by adding the cleanliness area corresponding to the cleanliness area when the elderly's clothes are just worn and the secondary cleanliness area corresponding to the cleanliness area on the nth day of wearing the elderly's clothes to obtain the overall cleanliness area; The method for obtaining the average sanitation value of the overall cleanliness area is: calculate the colorless average value corresponding to the cleanliness area when the elderly person's clothes are just worn and the secondary colorless average value corresponding to the cleanliness area when the elderly person's clothes are worn for the nth day, and use the average value as the sanitation average value of the entire cleanliness area.

7. A method for managing elderly care services according to claim 6, characterized in that: In step A005, the situation information includes dust quantity, humidity value, and temperature value.

8. A method for managing elderly care services according to claim 7, characterized in that: In step A005, a training set is established for each situation information and the overall cleanliness area and hygiene average value of the elderly's clothes in the corresponding situation, and multiple training sets are obtained. The training sets are used to perform TCN planning network training. The specific operation steps of performing TCN planning network training are as follows: A0051: Send the information about the elderly's clothes from the time they are first worn to the nth day to the TCN planning network, and obtain the average hygiene values ​​corresponding to the overall hygiene zones and overall cleanliness zones of the elderly's clothes in the next time period; A0052: The convolution structure shape and size of the TCN planning network is: [2,4]. The encoding device is used to obtain the associated characteristic data within a period of time, and the overall hygiene partition of the elderly's clothes in the next time period and the average hygiene value corresponding to the overall cleanliness area are obtained through the associated network.

9. A method for managing elderly care services according to claim 8, characterized in that: In step A006, the real-time situation information of the clothes worn by the elderly in the next time period is obtained, and input into the TCN planning network for training, so as to obtain the regional area estimation value and the average hygiene estimation value of the overall cleanliness of the elderly's clothes in the next time period; According to the estimated value of the regional area and the estimated value of the average hygiene level, the cleanliness of the elderly's clothes in the next time period is calculated, that is, B=C+F, wherein B is the cleanliness level, C is the estimated value of the regional area, and F is the estimated value of the average hygiene level. When the cleanliness level is lower than the minimum cleanliness level, a prompt is turned on to remind the staff to wash the elderly's clothes in time.

10. A pension service management system, characterized by: The elderly care service management method described in any one of claims 1 to 9 is applied to an elderly care service management system.

Images