Nursing service supervision method based on the fusion of Beidou satellite data and AI graphics

Abstract

The present invention proposes a nursing service supervision method based on the fusion of Beidou satellite data and AI graphics, which relates to the technical field of nursing service supervision. Based on Beidou satellite positioning, multiple nursing stations are obtained, and positioning data of the multiple nursing stations are assigned a positioning number to each nursing station; video data of each nursing station with a positioning number is collected in real time, and multiple video frame images in the video data of a fixed time period are extracted; multiple video frame images in each fixed time period are imported into a judgment unit of an AI processor, and multiple video frame images with similarities are extracted to form an image set; the image set is sent to a recognition model of the AI ​​processor, and each image in the image set is compared with a standard template image of a corresponding time period, and an unmatched image set and its corresponding time period and a nursing station positioning number are output.

Description

Technical Field

[0001] The present invention proposes a nursing service supervision method based on the fusion of Beidou satellite data and AI graphics, which relates to the technical field of nursing service supervision. Background Art

[0002] Nursing service supervision requires the use of various technical equipment, such as remote monitoring equipment, sensors, etc. Although these devices are powerful, they may not be used correctly by the nursing objects or caregivers due to complex operations or too many nursing objects. This may not only affect the accuracy and timeliness of the data, but also reduce the overall quality of nursing services. The integration of Internet, artificial intelligence and Internet of Things technologies is profoundly changing the face of the nursing field. These technologies not only improve the efficiency of data collection and use in the nursing field, but also optimize the nursing process, improve the quality of nursing, and provide new possibilities for remote nursing. How to monitor and evaluate each link in the nursing service process in real time, find problems in time and make improvements is the main problem facing the nursing field.

[0003] The progress of AI technology in remote patient care enables caregivers to use digital monitoring to track health data, identify abnormalities, and provide care without face-to-face supervision. The application of AI in the field of nursing provides direction for analysis results through enhanced intelligent assistance technology, while playing an important role in quality control in the process of providing services. While remote monitoring technology provides a large amount of patient data, how to effectively analyze and utilize this data remains a challenge. Summary of the invention

[0004] In order to solve the above technical problems, the present invention proposes a nursing service supervision method based on the fusion of Beidou satellite data and AI graphics, comprising the following steps:

[0005] Step 1: Locate multiple nursing stations based on Beidou satellites, obtain positioning data of multiple nursing stations, and assign a positioning number to each nursing station;

[0006] Step 2: Collect video data of each nursing station with a positioning number in real time, and extract multiple video frame images within a fixed time period in the video data;

[0007] Step 3: import multiple video frame images within each fixed time period into the judgment unit of the AI ​​processor, extract multiple video frame images with similarity within each fixed time period, and form an image set;

[0008] Step 4: Send the image set to the recognition model of the AI ​​processor, compare each image in the image set with the standard template image of the corresponding time period, and output the unmatched image set and its corresponding time period as well as the nursing station location number.

[0009] Furthermore, in step 3, the judgment unit calculates the correlation value M between the directional gradient values ​​of all adjacent video frame images within the same fixed time period:

[0010] , ;

[0011] Among them, i and j are the indexes of adjacent video frame images, L is the number of video frame images in the same fixed time period, and G i , G j is the directional gradient value of the i-th and j-th video frame images, and outputs multiple video frame images whose correlation value M is greater than the similarity threshold, forming an image set Y={Y 1 , …, Y k , …, Y S},Y k represents the kth image, and the number of images in the image set is S.

[0012] Furthermore, in step 4, the image set Y is fed into the recognition model, and each image in the image set Y is compared with the standard template image h of the corresponding time period I. I After comparison, the identification model is:

[0013] ;

[0014] In the formula: P hI represents the histogram feature mean of the standard template image in time period I; P Yk represents the kth image Y in the image set Y k The histogram feature mean, L I represents the output value of the video recognition model for time period I. If the video recognition model output value L I If it is not higher than the recognition threshold, the time period I that does not conform to the standard template image is output.

[0015] Furthermore, in the step 4, a mean image of multiple images in the image set is obtained, and the row pixel value means of the mean image are compared with the corresponding row pixel value means in the standard template image:

[0016] (1);

[0017] Among them, H a is the sum of the grayscale values ​​of the pixels in the ath row of the mean image, A is the total number of pixel rows in the mean image, and H b is the sum of the grayscale values ​​of the pixels in the bth row of the standard template image, B is the total number of pixel rows in the standard template image, is the row threshold;

[0018] Compare the mean values ​​of the columns of the mean image with the mean values ​​of the corresponding columns of the standard template image:

[0019] (2);

[0020] Among them, C q is the sum of the grayscale values ​​of the qth column of the mean image, Q is the total number of pixel columns in the mean image, and C t is the sum of the grayscale values ​​of the pixels in the tth column of the standard template image, T is the total number of pixel columns in the standard template image, is the column threshold;

[0021] When the mean values ​​of the row and column pixel values ​​of the mean image all meet the above conditions (1) and (2), the mean image

[0022] Like conforming to the standard template image.

[0023] Furthermore, in step one, the nursing station is spatially positioned through the Beidou satellite system to obtain positioning data of multiple nursing stations; after the positioning data is encrypted, it is sent to the central server through the transmission unit, and the central server binds the positioning data of each nursing station with the video data of the corresponding nursing station, and assigns a positioning number to the video data of each nursing station.

[0024] Furthermore, when the AI ​​processor outputs an unmatched set of images and their corresponding time periods and the nursing station location number, it proves that the nursing staff has not arrived at the designated location to perform the task at the scheduled time, and a reminder and alarm are issued.

[0025] Furthermore, the directional gradient value G of the video frame image is calculated:

[0026] ;

[0027] , ;

[0028] Among them, G X is the gradient value in the X direction, G Y is the gradient value in the Y direction, Sobel X is the gradient factor in the X direction, Sobel Y is the gradient factor in the Y direction, and Z is the pixel value of the pixel in the video frame image.

[0029] Furthermore, the gradient factor Sobel in the X direction X and the gradient factor Sobel in the Y direction Y They are:

[0030] ;

[0031] .

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

[0033] A nursing service supervision method based on the fusion of Beidou satellite data and AI graphics, which relates to the technical field of nursing service supervision, locates multiple nursing stations based on Beidou satellites, obtains the positioning data of multiple nursing stations, and assigns a positioning number to each nursing station; collects video data of each nursing station with a positioning number in real time, extracts multiple video frame images in the video data of a fixed time period; imports multiple video frame images in each fixed time period into the judgment unit of an AI processor, extracts multiple video frame images with similarity, and forms an image set; sends the image set to the recognition model of the AI ​​processor, compares each image of the image set with the standard template image of the corresponding time period, outputs the unmatched image set and its corresponding time period and the positioning number of the nursing station, and can monitor and evaluate each link in the nursing service process in real time, find problems in time and make improvements, thereby improving the overall quality of nursing services. Nursing staff can use digital monitoring tools to remotely track patients' health data, including vital signs, drug use, etc. The real-time monitoring function ensures the continuity and timeliness of data, which helps medical staff to find and deal with abnormal situations in time. BRIEF DESCRIPTION OF THE DRAWINGS

[0034] In order to more clearly illustrate the technical solutions in the embodiments of the present invention, the accompanying drawings required for use in the description of the embodiments will be briefly introduced below. Obviously, the accompanying drawings in the following description are only some embodiments of the present invention. For ordinary technicians in this field, other accompanying drawings can be obtained based on these accompanying drawings without paying creative labor.

[0035] Figure 1 It is a flow chart of the nursing service supervision method based on the fusion of Beidou satellite data and AI graphics of the present invention;

[0036] Figure 2 A schematic diagram of multiple nursing stations and a central server of the present invention;

[0037] Figure 3 Schematic diagram of the AI ​​processor structure of the present invention. DETAILED DESCRIPTION

[0038] In order to make the purpose, technical solution and advantages of the embodiments of the present application clearer, the technical solution in the embodiments of the present application will be clearly and completely described below in conjunction with the drawings in the embodiments of the present application. Obviously, the described embodiments are part of the embodiments of the present application, not all of the embodiments. Based on the embodiments in the present application, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of this application.

[0039] In the drawings of the specific embodiments of the present invention, in order to better and more clearly describe the working principles of the various components in the system, the connection relationship of the various parts in the device is shown, which only clearly distinguishes the relative position relationship between the various components, and cannot constitute a limitation on the signal transmission direction, connection sequence and structural size, size and shape of each part within the component or structure.

[0040] like Figure 1 The figure shows a flow chart of a nursing service supervision method based on the fusion of Beidou satellite data and AI graphics of the present invention, and the nursing service supervision method comprises the following steps:

[0041] Step 1: Locate multiple nursing stations based on Beidou satellites, obtain positioning data of multiple nursing stations, and assign a positioning number to each nursing station.

[0042] The Beidou satellite system is used to locate the nursing stations in space and obtain the location data of multiple nursing stations. After the location data is encrypted, it is sent to the central server through the transmission unit. Figure 2 As shown, the central server binds the positioning data of each nursing station with the uploaded data of the corresponding nursing station, and assigns a positioning number to the uploaded data of each nursing station.

[0043] Each nursing station also receives signals from the Beidou satellite system through a receiving module to determine its own location. The positioning data includes the latitude and longitude coordinates of the nursing station.

[0044] Each nursing station is assigned a unique positioning number, which can be associated with the positioning data of the nursing station (such as longitude and latitude coordinates) to facilitate subsequent data management and query.

[0045] Install a video surveillance system at each nursing station to collect video data from the nursing station. These video data include real-time surveillance video, playback video, etc. Add a location number to each video data file by including the location number in the file name or adding relevant information in the metadata of the video file.

[0046] The video data with the location number is stored in the server to ensure the security and accessibility of the data. At the same time, a corresponding data management system is established to facilitate the query, retrieval and playback of video data.

[0047] Step 2: Collect video data of each nursing station with a positioning number in real time, and extract multiple video frame images from the video data in a fixed time period.

[0048] Fixed time periods refer to fixed inspection times set according to nursing rules, during which nursing staff conduct daily inspections and status records of instruments or patients to ensure that the care recipients receive regular and systematic care, that the instruments operate normally, and that data is recorded in a timely manner.

[0049] Specifically, at a fixed time period every day, nursing staff will operate instruments to monitor the heart rate, blood pressure, body temperature, respiratory rate, etc. of the person being cared for; and ensure that the patient takes medication on time.

[0050] For care recipients who use testing instruments, such as ventilators and ECG monitors, caregivers need to check the instrument's functions and the patient's responses at fixed time intervals.

[0051] Recording and Documentation: After each nursing activity, the nursing staff needs to record the patient's health status, nursing measures and any abnormalities in the recording location so that doctors and other nursing staff can understand the patient's latest condition.

[0052] Video frame images at multiple time points in the video data within a fixed time period are extracted, and the multiple video frame images are preprocessed. The preprocessing includes zero point correction, background elimination and filtering. At the same time, denoising, transformation and smoothing operations are taken to enhance the graphic features of the image.

[0053] Step 3: Import multiple video frame images within each fixed time period into the judgment unit of the AI ​​processor, judge multiple video frame images with similarities, and form an image set.

[0054] The L video frame images in each fixed time period are imported into the judgment unit of the AI ​​graphics processor. The AI ​​processor uses the judgment unit to calculate the directional gradient value of each video frame image, judges the similarity between the video frame images through the similarity function f, and calculates the correlation value M between the directional gradient values ​​of all adjacent video frame images in the same time period.

[0055] , ;

[0056] Among them, i and j are the indexes of adjacent video frame images, L is the number of video frame images in a fixed time period, and G i , G j is the directional gradient value of the i-th and j-th video frame images, and outputs multiple video frame images whose correlation value M is greater than the similarity threshold, and obtains video frame images that meet the subsequent recognition conditions, that is, forms an image set Y={Y 1 , …, Yk , …, Y S},Y k Indicates the kth image. The number of images S in the image set is less than L, that is, among the L video frame images, only S video frame images meet the subsequent recognition conditions and can be input into the next recognition model.

[0057] The directional gradient value of the video frame image is calculated by calculating the gradient magnitude and direction in the x and y directions for each pixel in the grayscale image of the video frame image. Specifically:

[0058] Get the gradient value G in the X direction of the video frame image X and the gradient value G in the Y direction Y :

[0059] , ;

[0060] Wherein, Z is the pixel value of the pixel point in the video frame image.

[0061] pass Get the final directional gradient value.

[0062] Sobel gradient factor in the X direction X and the gradient factor Sobel in the Y direction Y They are:

[0063] ;

[0064] .

[0065] Step 4: Send the image set to the recognition model of the AI ​​processor, compare each image in the image set with the standard template image of the corresponding time period, and output the unmatched image set and its corresponding time period as well as the nursing station location number.

[0066] Standard template images are standard nursing posture or nursing position images pre-taken at fixed time periods for comparison with the image set to be compared. Standard template images are high-definition, standardized standard template image sets captured at different time points for specific nursing operations or patient postures, planned and strictly executed according to the nursing process steps.

[0067] The image set Y is sent to the recognition model, and each image in the image set Y is compared with the standard template image h of the corresponding time period I. I The system compares and outputs the time periods that do not conform to the standard template images, thereby identifying the non-compliant operation links in the nursing process, thereby achieving real-time monitoring and evaluation of the quality of nursing services. For example, at 10 a.m., the patient fails to arrive at the designated location to check the operation of the equipment as required.

[0068] The recognition model focuses on evaluating whether the nursing operation meets the requirements by comparing the standard template image of the nursing operation in time period I with each image in the image set.

[0069] The calculation formula of the recognition model is:

[0070] ;

[0071] In the formula: P hI represents the histogram feature mean of the standard feature template in time period I; P Yk represents the kth image Y k In order to calculate the above two histogram feature means, it is necessary to first convert the standard feature template and each image in the image set into a histogram form, and then calculate the histogram feature mean of each histogram.

[0072] In the formula: L I Represents the output value of the video recognition model, the output value of the video recognition model L I If it is higher than the recognition threshold, it proves that the caregiver in time period I arrives at the designated location to perform nursing operations as required and meets the requirements. Otherwise, it does not meet the requirements, and the positioning data and time period of the nursing station that does not meet the requirements and the corresponding video data are output.

[0073] Based on the above supervision methods, develop detailed standard operating procedures for each nursing task, including the specific time, location, equipment and materials required for the task. Record and track: Use electronic systems or logs to record the time and place where the nursing staff performs the task, and ensure that all operations are recorded. Set reminders and alarms: Set reminders and alarms in the system, and the system will sound an alarm when the nursing staff does not arrive at the designated location or perform the task at the scheduled time. Through the real-time monitoring system, managers can immediately detect and correct non-compliant operations. Feedback the monitoring results to nursing staff and management to improve the operating process and service quality in a timely manner.

[0074] Example 1

[0075] The recognition model focuses on evaluating whether the nursing operation meets the requirements by comparing the standard template image of the nursing operation in time period I with each image in the image set. The recognition model can also be constructed in the following way.

[0076] Get the mean image of multiple images in the image set, and compare the rows and columns of the mean image with the rows and columns corresponding to the standard template image, specifically:

[0077] The row pixel values ​​of the mean image are compared with the corresponding row pixel values ​​in the standard template image. The comparison formula is:

[0078] (1);

[0079] Among them, H a is the sum of the grayscale values ​​of the pixels in the ath row of the mean image, A is the total number of pixel rows in the mean image, and H b is the sum of the grayscale values ​​of the pixels in the bth row of the standard template image, B is the total number of pixel rows in the standard template image, is the row threshold.

[0080] The mean values ​​of the columns of the mean image are compared with the corresponding mean values ​​of the columns of the standard template image. The comparison formula is:

[0081] (2);

[0082] Among them, C q is the sum of the grayscale values ​​of the qth column of the mean image, Q is the total number of pixel columns in the mean image, and C t is the sum of the grayscale values ​​of the pixels in the tth column of the standard template image, T is the total number of pixel columns in the standard template image, is the column threshold.

[0083] When the mean values ​​of the row and column pixel values ​​of the mean image all meet the above conditions (1) and (2), the mean image meets the standard template image. This proves that the caregiver in the time period corresponding to the mean image arrives at the designated location to perform nursing operations as required, which meets the requirements. Otherwise, when the mean values ​​of the row and column pixel values ​​of the mean image do not meet the above conditions (1) and (2), it proves that the mean image does not meet the requirements, and the positioning data and time period of the nursing station that does not meet the requirements and the corresponding video data are output at the same time.

[0084] Example 2

[0085] The present invention also constructs a nursing service supervision system using a nursing service supervision method based on the fusion of Beidou satellite data and AI graphics, including: a Beidou satellite system, a central server, a transmission unit, a video frame extraction unit, and an AI processor.

[0086] The BeiDou satellite system provides accurate time and location data to ensure that the supervision of nursing services is highly accurate and reliable. The BeiDou satellite system is used to locate the nursing stations in space and obtain the positioning data of multiple nursing stations. The positioning data is encrypted and sent to the central server through the transmission unit. The central server, as the data processing and storage center of the system, is responsible for receiving, processing and storing videos from each nursing station. The central server binds the positioning data of each nursing station with the video data of the corresponding nursing station and assigns a positioning number to the video data of each nursing station.

[0087] The video data of each nursing station with a positioning number is collected, and a plurality of video frame images in the video data of a fixed time period are extracted by a video frame extraction unit.

[0088] like Figure 3 As shown, the AI ​​processor includes: a judgment unit, a recognition model, and an output unit. The judgment unit judges whether the nursing service meets the predetermined standards and requirements based on the Beidou satellite data and the data extracted from the video frame. The recognition model identifies whether the caregiver arrives at the designated location to perform the nursing operation as required. The output unit outputs the corresponding supervision information based on the results of the recognition model. If the nursing service meets the requirements, confirmation information may be output; if not, the positioning data, time period and corresponding video data of the nursing station that do not meet the requirements are output.

[0089] Specifically, a plurality of video frame images within each fixed time period are imported into the judgment unit of the AI ​​processor. Specifically, the judgment unit forms an image set by judging a plurality of video frame images having similarities.

[0090] The image set is sent to the recognition model of the AI ​​processor, and the recognition model compares each image in the image set with the standard template image of the corresponding time period. The output unit outputs the unmatched image set and its corresponding time period as well as the nursing station location number.

[0091] By combining Beidou satellite data, the present invention and AI graphics fusion technology can conduct in-depth analysis of the large amount of nursing service data collected, and through image recognition and data processing technology, achieve effective supervision of the quality of nursing services.

[0092] In the above embodiments, it can be implemented in whole or in part by software, hardware, firmware or any combination thereof. When implemented using software, it can be implemented in whole or in part in the form of a computer program product. The computer program product includes one or more computer instructions. When the computer program instructions are loaded and executed on a computer, the process or function described in the embodiment of the present application is generated in whole or in part. The computer may be a general-purpose computer, a special-purpose computer, a computer network, or other programmable device. The computer instructions may be stored in a computer-readable storage medium or transmitted through the computer-readable storage medium. The computer-readable storage medium may be any available medium that a computer can access or a data storage device such as a server or a data center that includes one or more available media integrated. The available medium may be a magnetic medium, (e.g., a floppy disk, a hard disk, a tape), an optical medium (e.g., a DVD), or a semiconductor medium (e.g., a solid state disk (SSD)), etc.

[0093] The above is only a specific implementation of the present application, but the protection scope of the present application is not limited thereto. Any technician familiar with the technical field can easily think of various equivalent modifications or replacements within the technical scope disclosed in the present application, and these modifications or replacements should be included in the protection scope of the present application. Therefore, the protection scope of the present application shall be based on the protection scope of the claims.

Claims

1. A nursing service supervision method based on the fusion of Beidou satellite data and AI graphics, characterized in that: The steps include: Step 1: Locate multiple nursing stations based on Beidou satellites, obtain positioning data of multiple nursing stations, and assign a positioning number to each nursing station; Step 2: Collect video data of each nursing station with a positioning number in real time, and extract multiple video frame images within a fixed time period in the video data; Step 3: import multiple video frame images within each fixed time period into the judgment unit of the AI ​​processor, extract multiple video frame images with similarity within each fixed time period, and form an image set; The judgment unit calculates the correlation value M between the directional gradient values ​​of all adjacent video frame images within the same fixed time period: ， ； Among them, i and j are the indexes of adjacent video frame images, L is the number of video frame images in the same fixed time period, and G i , G j is the directional gradient value of the i-th and j-th video frame images, and outputs multiple video frame images whose correlation value M is greater than the similarity threshold, forming an image set Y={Y1, …, Y k , …, Y S },Y k represents the kth image, and the number of images in the image set is S; Step 4: Send the image set to the recognition model of the AI ​​processor, compare each image in the image set with the standard template image of the corresponding time period, and output the unmatched image set and its corresponding time period and the nursing station location number; The image set is sent to the recognition model, and each image in the image set is compared with the standard template image of the corresponding time period. The recognition model is: ； In the formula: P hI The standard template image h represents the time period I I The histogram feature mean of Yk represents the kth image Y in the image set Y k The histogram feature mean, L I represents the output value of the video recognition model for time period I. If the video recognition model output value L I If it is not higher than the recognition threshold, the time period I that does not conform to the standard template image is output.

2. The nursing service supervision method according to claim 1, characterized in that: In step 4, the mean image of multiple images in the image set is obtained, and the row pixel value means of the mean image are compared with the corresponding row pixel value means in the standard template image: （1） Among them, H a is the sum of the grayscale values ​​of the pixels in the ath row of the mean image, A is the total number of pixel rows in the mean image, and H b is the sum of the grayscale values ​​of the pixels in the bth row of the standard template image, B is the total number of pixel rows in the standard template image, is the row threshold; Compare the mean values ​​of the columns of the mean image with the mean values ​​of the corresponding columns of the standard template image: （2） Among them, C q is the sum of the grayscale values ​​of the qth column of the mean image, Q is the total number of pixel columns in the mean image, and C t is the sum of the grayscale values ​​of the pixels in the tth column of the standard template image, T is the total number of pixel columns in the standard template image, is the column threshold; When the row and column pixel value means of the mean image all meet the above conditions (1) and (2), the mean image conforms to the standard template image.

3. The nursing service supervision method according to claim 1, characterized in that: In step one, the nursing station is spatially positioned through the Beidou satellite system to obtain the positioning data of multiple nursing stations; after the positioning data is encrypted, it is sent to the central server through the transmission unit. The central server binds the positioning data of each nursing station with the video data of the corresponding nursing station and assigns a positioning number to the video data of each nursing station.

4. The nursing service supervision method according to claim 1, characterized in that: When the AI ​​processor outputs an unmatched set of images and their corresponding time periods and the nursing station location number, it proves that the nursing staff did not arrive at the designated location to perform the task at the scheduled time, and issues a reminder and alarm.

5. The nursing service supervision method according to claim 1, characterized in that: Calculate the directional gradient value G of the video frame image: ； ， ； Among them, G X , G Y is the gradient value in the X and Y directions, Sobel X and Sobel Y is the gradient factor in the X and Y directions, and Z is the pixel value of the pixel in the video frame image.

6. The nursing service supervision method according to claim 5, characterized in that: Sobel gradient factor in X and Y directions X and Sobel Y for: ； 。

Images