A remote monitoring and management method for postpartum rehabilitation of a parturient

Abstract

The present application relates to the field of maternal body image encoding, and particularly relates to a maternal postpartum rehabilitation remote monitoring and management method. According to the image change of the adjacent two frames of the first frame image per second, the motion coefficient is obtained, and then the dynamic and static interval frame image is selected. According to the edge distribution between the adjacent two frames of each frame image, the motion characteristic value is obtained. The maternal body images are clustered to obtain the maternal body image cluster. The reference coefficient is obtained according to the edge distribution and information entropy in the image. Each image cluster is predicted and encoded according to the reference coefficient of each frame image in each image cluster. The present application can reduce the calculation cost, improve the efficiency of the monitoring video compression, and enable the relevant personnel to timely perform the remote monitoring and management on the postpartum maternal.

Description

TECHNICAL FIELD

[0001] The present application relates to the field of maternal body image encoding, in particular to a maternal postpartum rehabilitation remote monitoring management method. BACKGROUND

[0002] Maternal postpartum rehabilitation remote monitoring is gradually becoming an important supplement to postpartum care. Existing remote monitoring methods mainly rely on intelligent wearable devices and sensors to monitor maternal vital sign data, but the limitations of these devices make it difficult to fully track maternal movement and rehabilitation activities. Using remote video monitoring for activity tracking can make up for this deficiency. Video monitoring can capture maternal movements and postures in real time, helping doctors accurately assess the recovery status of the mother, especially in the early postpartum period. By observing the mother's activities, doctors can promptly identify adverse postures or improper movements, thereby avoiding problems such as back pain or pelvic floor muscle dysfunction. In addition, remote video monitoring can also provide personalized rehabilitation guidance to ensure that the mother can perform scientific and effective rehabilitation training at home, reducing unnecessary hospital visits, reducing the mother's stress, and improving rehabilitation effectiveness.

[0003] Applying predictive coding to video images not only helps reduce bandwidth usage when transmitting real-time video streams, but also optimizes storage and subsequent playback. When it is necessary to play back remote monitoring videos, compressed files are smaller, easy to quickly search and play, while still maintaining high video quality, which is very helpful for monitoring and evaluating the rehabilitation process of the mother. However, traditional encoding methods require motion estimation for each frame to find the optimal reference frame. This motion estimation process is computationally intensive for long video streams or high-resolution videos. During the postpartum rehabilitation process, the mother will produce frequent changes in action patterns such as sitting- standing transitions and walking. If the process of finding a reference frame and calculating motion vectors is repeated each time, it will increase the computational complexity and reduce the transmission efficiency, thereby making it difficult for relevant personnel to timely monitor and manage the mother's postpartum rehabilitation remotely. SUMMARY

[0004] In order to solve the technical problem that during the predictive coding of maternal body images, the mother will produce frequent changes in action patterns such as sitting- standing transitions and walking during the rehabilitation process, and if the process of finding a reference frame and calculating motion vectors is repeated each time, it will increase the computational complexity and reduce the transmission efficiency, thereby making it difficult for relevant personnel to timely monitor and manage the mother's postpartum rehabilitation remotely, the purpose of the present application is to provide a maternal postpartum rehabilitation remote monitoring management method, the technical scheme adopted is as follows:

[0005] A maternal postpartum rehabilitation remote monitoring management method, the method comprising:

[0006] Collecting continuous frames of maternal body images at different sampling time points after childbirth; taking the first frame of maternal body image obtained at each sampling time point as the first frame of maternal image at each sampling time point;

[0007] Optionally, the first frame of maternal image at one sampling time point is taken as a reference time image; according to the image change characteristics between the adjacent two frames of maternal body images of the reference time image, the motion coefficient of the reference time image is obtained; according to the motion coefficient, all the first frames of maternal images are screened to obtain all the dynamic and static interval frame images; optionally, one frame of maternal body image is taken as a reference frame image; according to the edge distribution difference between the adjacent two frames of maternal body images of the reference frame image, and the edge shape feature and the edge distribution feature in the reference frame image, the motion feature value of the reference frame image is obtained; according to the motion feature value, all the frames of maternal body images in each adjacent two dynamic and static interval frame images are clustered to obtain all the clusters of maternal body images; according to the edge distribution feature and the information entropy in each frame of maternal body image, the reference coefficient of each frame of maternal body image is obtained;

[0008] According to the reference coefficient of each frame of maternal body image in each cluster of maternal body images, each cluster of maternal body images is predicted and encoded.

[0009] Further, the method for obtaining the motion coefficient comprises:

[0010] According to the frame difference method, the difference area between the adjacent two frames of maternal body images of the reference time image is obtained;

[0011] According to the motion coefficient calculation formula, the motion coefficient is obtained, and the motion coefficient calculation formula is as follows:

[0012] ;

[0013] In the formula, The motion coefficient of the reference time image is represented; The number of pixel points in the difference area between the previous frame of maternal body image of the reference time image and the reference time image is represented; The number of pixel points in the difference area between the next frame of maternal body image of the reference time image and the reference time image is represented; The information entropy in the reference time image is represented; The information entropy of the previous frame of maternal body image of the reference time image is represented; The information entropy of the next frame of maternal body image of the reference time image is represented; The maximum value function is represented.

[0014] Further, the method for obtaining the dynamic and static interval frame image comprises:

[0015] The first frame of the mother's image with a motion coefficient greater than a preset first threshold is used as the static-dynamic interval frame image.

[0016] Furthermore, the method for obtaining the motion feature values ​​includes:

[0017] The motion feature value is obtained according to the motion feature value calculation formula, which is as follows:

[0018] ;

[0019] In the formula, Represents the motion feature values ​​of the reference frame image; Indicates the number of edges in the reference frame image; This indicates the number of edges in the previous frame of the mother's body image in the reference frame image; This indicates the number of edges in the next frame of the mother's body image following the reference frame image. Indicates the reference frame image with the first The preset number of other edges that are closest to each edge; Indicates the reference frame image with the first The nearest edge distance is the first The curvature of the other edges; Indicates the reference frame image with the first The nearest edge distance is the first The number of pixels in each of the other edges; Indicates the reference frame image with the first The nearest edge distance is the first The distance between the endpoints of the other edges; Indicates the first The edge and the nearest one The distance between the other edges; This represents the absolute value function.

[0020] Furthermore, the method for acquiring the maternal body image cluster includes:

[0021] Calculate the mean motion feature value of all frames of maternal body images in each pair of adjacent dynamic and static interval frames to obtain the first mean; use the first mean to cluster all maternal body images into dynamic maternal image clusters and static maternal image clusters.

[0022] The dynamic and static maternal image clusters are clustered twice using the motion feature values ​​of each frame of maternal body image to obtain all maternal body image clusters.

[0023] Furthermore, the method for obtaining the reference coefficient includes:

[0024] Calculate the information entropy of all other pixel points in the preset neighborhood of each pixel point in each frame of maternal body image as the reference information entropy of each pixel point; and accumulate and sum the reference information entropy of all pixel points constituting the edge in the reference maternal body image cluster to obtain the reference coefficient of each frame of maternal body image.

[0025] Further, the prediction encoding of each maternal body image cluster according to the reference coefficient of each frame of maternal body image in each maternal body image cluster comprises:

[0026] Optionally, one maternal body image cluster is taken as the reference maternal body image cluster; and the maternal body image with the maximum reference coefficient in the reference maternal body image cluster is taken as the target image;

[0027] Calculate the difference between the target image and each other maternal body image in the reference maternal body image cluster to obtain the residual image corresponding to each other maternal body image in the reference maternal body image cluster; and quantitatively encode all the residual images to obtain the encoding result of the reference maternal body image cluster;

[0028] Traverse each maternal body image cluster to obtain the encoding result of each maternal body image cluster.

[0029] A maternal postpartum rehabilitation remote monitoring management system, the system comprising a memory, a processor and a computer program stored in the memory and executable on the processor, the processor implementing the steps of the above maternal postpartum rehabilitation remote monitoring management method when executing the computer program.

[0030] A computer readable storage medium, the computer readable storage medium storing a computer program, the computer program being executed by a processor to implement the steps of the above maternal postpartum rehabilitation remote monitoring management method.

[0031] A computer device comprising a memory, a processor and a computer program stored in the memory and executable on the processor, the processor implementing the steps of the above maternal postpartum rehabilitation remote monitoring management method when executing the computer program.

[0032] The present application has the following advantages:

[0033] The application collects continuous frame maternal body images at different sampling time points, when maternal movement causes image blur, it can be considered that the maternal keeps the same movement state within one second at this time, therefore, the first frame maternal body image obtained at each sampling time point is regarded as the first frame maternal image at each sampling time point, the first frame maternal image is used to reflect the maternal movement feature of each second, and then the dynamic and static interval frame image is positioned; when the picture of the monitoring video changes from static to dynamic, the movement of the maternal will cover the originally static part of the scene, meanwhile, the maternal movement exposes the part of the scene which was covered before the movement, therefore, there is a large image change in the front and back two frame maternal body images of the dynamic and static interval frame image, therefore, the adjacent two frame maternal body images of the first frame maternal image are analyzed to obtain the movement coefficient; the dynamic and static interval frame image is screened by using the movement coefficient; since the static image has no obvious dynamic blur or movement trace, the background and the edge of the character change little, therefore, the edge distribution difference between the adjacent two frame maternal body images of the reference frame image, and the edge shape feature and the edge distribution feature in the reference frame image are used to obtain the movement feature value of the reference frame image, the static image and the dynamic image are separated according to the movement feature value; since the traditional prediction coding method is difficult to select the best maternal body image as the reference frame for subsequent coding operation, the movement feature values of different maternal body images are used to cluster the static image and the dynamic image to obtain several maternal body image clusters; subsequently, the best maternal body image in the reference maternal body image cluster is found, and then the prediction coding is performed on each maternal body image cluster. The application can reduce the calculation cost, improve the efficiency of monitoring video compression, and enable relevant personnel to remotely monitor and manage the maternal after delivery. BRIEF DESCRIPTION OF DRAWINGS

[0034] In order to more clearly illustrate the technical solutions in the embodiments of the present application or the prior art, and the advantages thereof, below, a brief introduction will be given to the drawings needed to be used in the embodiments or the prior art description. Obviously, the drawings in the following description only show some embodiments of the present application, and for those skilled in the art, other drawings can be obtained without creative effort based on these drawings.

[0035] Figure 1 A flow chart of a maternal postpartum rehabilitation remote monitoring and management method provided by an embodiment of the present application. DETAILED DESCRIPTION

[0036] In order to further clarify the technical means and effects taken by the present application to achieve the predetermined inventive purpose, the following describes in detail the specific implementation, structure, features and effects of a postpartum rehabilitation remote monitoring and management method for puerpera according to the present application in combination with the accompanying drawings and preferred embodiments. In the following description, different "one embodiment" or "another embodiment" do not necessarily refer to the same embodiment. In addition, the specific features, structures or characteristics in one or more embodiments can be combined in any suitable form.

[0037] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which the present application belongs.

[0038] The specific scheme of the postpartum rehabilitation remote monitoring and management method for puerpera provided by the present application is specifically described below in combination with the accompanying drawings.

[0039] Please refer to Figure 1 which shows a postpartum rehabilitation remote monitoring and management method for puerpera provided by an embodiment of the present application, which comprises:

[0040] Step S1: Collecting continuous frame puerpera body images at different sampling time points after delivery; taking the first frame puerpera body image obtained at each sampling time point as the first frame puerpera image at each sampling time point.

[0041] The present embodiment is mainly applied to the compression encoding scene of postpartum monitoring video of puerpera; in order to monitor and manage the rehabilitation state of puerpera after delivery, the monitoring video of puerpera during the rehabilitation period needs to be first shot, and the monitoring video is usually compressed due to the large amount of data of the video. In the present embodiment, the method of predictive encoding is usually used to compress the monitoring video, and the predictive encoding method is to perform motion estimation on each frame image of the monitoring video so as to find the optimal frame image, so in the present embodiment, the processing tool is used to extract frame images from the monitoring video, so as to collect continuous frame puerpera body images at different sampling time points after delivery.

[0042] In an embodiment of the present application, the sampling time is set to 1 second, that is, each frame puerpera body image in each second of the monitoring video is collected, and the frame rate of the monitoring video is set to 30 frames per second. It should be noted that the sampling time and the frame rate can be set by the user and are not limited herein.

[0043] ​Since in the continuous frame maternal body images, blurred images caused by maternal movement will appear, that is, dynamic images in the maternal body images, and the maternal body images without the motion subject blur are static images, and the difference between the static images and the dynamic images is very obvious, in order to subsequently find the dynamic and static interval frame images distinguishing the static images and the dynamic images, and thus perform predictive coding, in the embodiment of the present application, when the maternal movement causes image blur, it can be considered that the maternal maintains the same movement state within one second at this time, therefore, the first frame maternal body image acquired at each sampling time is taken as the first frame maternal image at each sampling time, the first frame maternal image is used to reflect the maternal movement features of each second, and then the dynamic and static interval frame images are positioned.

[0044] Step S2: optionally taking the first frame maternal image at one sampling time as a reference time image; obtaining a motion coefficient of the reference time image according to the image change features between the adjacent two frame maternal body images of the reference time image; screening all the first frame maternal images according to the motion coefficient, to obtain all the dynamic and static interval frame images; optionally taking one frame maternal body image as a reference frame image; obtaining a motion feature value of the reference frame image according to the edge distribution difference between the adjacent two frame maternal body images of the reference frame image, and the edge shape features and the edge distribution features in the reference frame image; clustering all the frame maternal body images in each adjacent two dynamic and static interval frame images according to the motion feature value, to obtain all the maternal body image clusters; and obtaining a reference coefficient of each frame maternal body image according to the edge distribution features and the information entropy in each frame maternal body image.

[0045] When the picture of the monitoring video changes from static to dynamic, the movement of the maternal will cover the originally static part of the scene, and meanwhile, the maternal movement will expose the part of the scene that was blocked before the movement, so in the front and back two frame maternal body images of the dynamic and static interval frame images, there will be a large image change, therefore, the first frame maternal image at one sampling time is optionally taken as the reference time image, in the embodiment of the present application, the motion coefficient of the reference time image is obtained according to the image change features between the adjacent two frame maternal body images of the reference time image.

[0046] Preferably, in the embodiment of the present application, the method for obtaining the motion coefficient comprises:

[0047] Since there will be a large image change in the front and back two frame maternal body images of the dynamic and static interval frame images, the difference area between the adjacent two frame maternal body images of the reference time image is obtained according to the frame difference method.

[0048] The motion coefficient is obtained according to the motion coefficient calculation formula, and the motion coefficient calculation formula is as follows:

[0049] ;

[0050] In the formula, Indicates the motion coefficients of the image at the reference time. This indicates the number of pixels in the difference region between the previous frame of the mother's body image and the reference time image. This represents the number of pixels in the difference region between the next frame of the mother's body image at the reference time and the reference time image. This represents the information entropy in the image at the reference time. The information entropy of the image of the mother's body in the previous frame of the reference time image; The entropy represents the information of the next frame of the mother's body image at the reference time. This represents the maximum value function.

[0051] In the formula for calculating motion coefficients, the more pixels in the difference region between two frames of the mother's body image before and after the reference time image, the higher the motion coefficient. The larger the value, the more significant the change in the mother's motion state between the two frames of the reference time image. In this case, the reference time image is more likely to be a frame with a motion-static interval. Since there will be a significant difference in the information entropy of images when the mother's body is static or in a static state, this... The larger the value, the more likely the reference time image is a static / dynamic interval frame image.

[0052] Preferably, in one embodiment of the present invention, the method for obtaining dynamic and static interval frame images includes:

[0053] The first frame of the mother's image with a motion coefficient greater than a preset first threshold is used as the motion-static interval frame image. In one embodiment of the present invention, the preset first threshold is set to 0.9. It should be noted that the preset first threshold can be set by itself and is not limited here.

[0054] If the postpartum body image between two static and dynamic frame images is a static image, then the texture in the postpartum body image will be relatively smooth, without obvious motion blur or motion traces, and the background and figure edges will change little. If the postpartum body image between two static and dynamic frame images is a dynamic image, then due to the movement of the figure or limb movements, there will be a large pixel difference between adjacent frames of the postpartum body image. In addition, during the postpartum recovery process, the movements generated during the process are usually relatively gentle to ensure safety and avoid excessive stress on the body. Moreover, many rehabilitation movements focus more on improving the postpartum woman's overall coordination ability, thus generating multi-directional limb movements or overall movements. These will produce obvious motion blur. Therefore, compared to the static state, the postpartum woman will show more obvious motion blur in the moving limbs when performing movements. Therefore, in this embodiment of the invention, the motion feature value of the reference frame image is obtained based on the edge distribution difference between two adjacent frames of the postpartum body image of the reference frame image, as well as the edge shape features and edge distribution features in the reference frame image. The static image and the dynamic image are separated based on the motion feature value.

[0055] Preferably, in one embodiment of the present invention, the method for obtaining motion feature values ​​includes:

[0056] The motion feature value is obtained according to the motion feature value calculation formula, which is shown below:

[0057] ;

[0058] In the formula, Represents the motion feature values ​​of the reference frame image; Indicates the number of edges in the reference frame image; This indicates the number of edges in the previous frame of the mother's body image in the reference frame image; This indicates the number of edges in the next frame of the mother's body image following the reference frame image. Indicates the reference frame image with the first The preset number of other edges that are closest to each edge; Indicates the reference frame image with the first The nearest edge distance is the first The curvature of the other edges; Indicates the reference frame image with the first The nearest edge distance is the first The number of pixels in each of the other edges; Indicates the reference frame image with the first The nearest edge distance is the first The distance between the endpoints of the other edges; Indicates the first The edge and the nearest one The distance between the other edges; represents an absolute value function.

[0059] In the motion characteristic value calculation formula, since there are motion blur areas in the dynamic image, compared with the static image, the number of edges is less, so the number of edges of the reference frame image is smaller, which means that the reference frame image is more likely to belong to the dynamic image, and the motion characteristic value of the reference frame image is larger; since there is no dynamic blur area in the static image, and the change amplitude is very small within a short time, so the number of edges of the image is large and stable, and the difference between the number of edges of the adjacent two frame maternal body images and the reference frame image is larger, which means that the reference frame image is more likely to belong to the dynamic image, and the motion characteristic value of the reference frame image is larger; is larger, which means that the number of pixel points constituting the first other edge is much larger than the minimum number of pixel points required for the length of the edge, which means that the edge has a significant width, and if the curvature of the first other edge in the reference frame image is smaller, which means that the edge is closer to a straight line, which means that it is more likely to belong to the dynamic blur area, which means that the reference frame image is more likely to belong to the dynamic image, and the motion characteristic value of the reference frame image is larger; is smaller, which means that the blur edge is more concentrated, which means that the blur area is larger, which means that the reference frame image is more likely to belong to the dynamic image, and the motion characteristic value of the reference frame image is larger.

[0060] Since the traditional prediction encoding method is difficult to select the best maternal body image as the reference frame for subsequent encoding operation, in the embodiment of the present application, the motion characteristic values of different maternal body images are used to cluster static images and dynamic images, to obtain a plurality of maternal body image clusters, and then each maternal body image cluster is prediction encoded, so as to improve the operation efficiency.

[0061] Preferably, in an embodiment of the present application, the method for obtaining the maternal body image cluster comprises:

[0062] calculating the mean value of the motion characteristic value of all frame maternal body images in each adjacent two dynamic and static interval frame images to obtain a first mean value; clustering all maternal body images by using the first mean value into a dynamic maternal image cluster and a static maternal image cluster; in the embodiment of the present application, the clustering method is adopted, and the number of clustering clusters is set to 2, so as to divide all maternal body images into a dynamic maternal image cluster and a static maternal image cluster. It should be noted that the clustering method is a technical means known to those skilled in the art, which will not be described here.

[0063] ​​The motion feature values ​​of each frame of the mother's body image are used to perform secondary clustering on the dynamic mother image cluster and the static mother image cluster to obtain all mother body image clusters. In one embodiment of the present invention, the elbow method is used to determine the optimal number of clusters for secondary clustering. It should be noted that the elbow method is a well-known technique in the art and will not be described in detail here.

[0064] After secondary clustering, the motion feature values ​​of the maternal body images within each maternal body image cluster are the most similar. Therefore, the best maternal body image in the reference maternal body image cluster is found, and the reference maternal body image cluster is predicted and encoded. Then, all maternal body image clusters are traversed to obtain the encoding result of each maternal body image cluster.

[0065] Preferably, in one embodiment of the present invention, the method for obtaining the reference coefficient includes:

[0066] Calculate the information entropy of the grayscale values ​​of all other pixels within a preset neighborhood of each pixel in each frame of the maternal body image, and use this as the reference information entropy for each pixel. Then, sum the reference information entropies of all pixels forming the edges in the reference maternal body image cluster to obtain the reference coefficient for each frame of the maternal body image. The formula for calculating the reference coefficient is as follows:

[0067] ;

[0068] In the formula, This represents the reference coefficient for each frame of the mother's body image; This indicates the number of edges in each frame of the mother's body image; Indicates the first The number of pixels contained in each edge; Indicates the first The edge of the first In one embodiment of the present invention, the information entropy of the grayscale values ​​of pixels within a preset neighborhood of a given pixel is defined as follows: Centered on 100 pixels A rectangular area.

[0069] In the formula for calculating the reference coefficient, the greater the gray value information entropy of each pixel within each edge of the maternal body image, the richer the information contained in the maternal body image, and the more suitable it is to be used as the best maternal body image for prediction coding. In this case, the reference coefficient of the maternal body image is greater.

[0070] Step S3: Perform predictive coding for each maternal body image cluster based on the reference coefficients of each frame of maternal body image within each maternal body image cluster.

[0071] Preferably, in one embodiment of the present invention, the specific steps include:

[0072] Optionally, one maternal body image cluster is selected as a reference maternal body image cluster; a maternal body image with the largest reference coefficient in the reference maternal body image cluster is selected as a target image; a difference between the target image and each other maternal body image in the reference maternal body image cluster is calculated to obtain a residual image corresponding to each other maternal body image in the reference maternal body image cluster; all the residual images are quantization-encoded to obtain an encoding result of the reference maternal body image cluster; each maternal body image cluster is traversed to obtain an encoding result of each maternal body image cluster, thereby compressing the video data and reducing the required bit rate for storage and transmission.

[0073] In summary, continuous frame maternal body images at different sampling time points after childbirth are collected; a first frame maternal body image obtained at each sampling time point is selected as a first frame maternal image at each sampling time point; a first frame maternal image at one sampling time point is selected as a reference time image; a motion coefficient of the reference time image is obtained according to an image change feature between adjacent two frame maternal body images of the reference time image; all interval frame images between static and dynamic are obtained by screening all the first frame maternal images according to the motion coefficient; a frame maternal body image is selected as a reference frame image; a motion feature value of the reference frame image is obtained according to an edge distribution difference between adjacent two frame maternal body images of the reference frame image and an edge shape feature and an edge distribution feature in the reference frame image; all maternal body image clusters are obtained by clustering all frame maternal body images in each adjacent two interval frame images between static and dynamic according to the motion feature value; a reference coefficient of each frame maternal body image is obtained according to an edge distribution feature and an information entropy in each frame maternal body image; each maternal body image cluster is predicted encoded according to the reference coefficient of each frame maternal body image in each maternal body image cluster.

[0074] An embodiment of the present application has a second object to provide a maternal postpartum rehabilitation remote monitoring and management system, which comprises a memory, a processor and a computer program, wherein the memory is used to store the corresponding computer program, the processor is used to run the corresponding computer program, and the computer program can realize the method described in steps S1-S3 when running in the processor.

[0075] An embodiment of the present application has a third object to provide a computer device, which comprises a memory, a processor and a computer program stored in the memory and capable of running on the processor, and the processor realizes the method described in steps S1-S3 when executing the computer program.

[0076] An embodiment of the present application has a fourth object to provide a computer readable storage medium, which stores a computer program, and the computer program realizes the method described in steps S1-S3 when executed by a processor.

[0077] It is to be noted that the sequential order of the above-described embodiments of the present application only for the purpose of description, but not the advantages and disadvantages of the embodiments. The processes depicted in the drawings do not necessarily require the specific order or continuous order shown to achieve the desired results. In some embodiments, multi-task processing and parallel processing are also possible or can be advantageous.

[0078] Each of the embodiments in the specification is described in a progressive manner, and the same or similar parts between the embodiments can be referred to each other. Each embodiment focuses on the difference from other embodiments.

Claims

1. A method for remote monitoring and management of postpartum recovery for mothers, characterized in that, The method includes: Collect consecutive frames of the mother's body at different sampling times after childbirth; use the first frame of the mother's body acquired at each sampling time as the first frame of the mother's body at each sampling time; Select the first frame of the maternal image at any sampling time as the reference time image; obtain the motion coefficient of the reference time image based on the image change characteristics between two adjacent maternal body images at the reference time image; filter all the first frame maternal images based on the motion coefficient to obtain all dynamic and static interval frame images; select one maternal body image as the reference frame image; obtain the motion feature value of the reference frame image based on the edge distribution difference between two adjacent maternal body images at the reference frame image, as well as the edge shape features and edge distribution features in the reference frame image; cluster all maternal body images in each pair of adjacent dynamic and static interval frame images based on the motion feature value to obtain all maternal body image clusters; obtain the reference coefficient of each maternal body image based on the edge distribution features and information entropy within each maternal body image. Predictive coding is performed on each of the maternal body image clusters based on the reference coefficients of each frame of maternal body image within each maternal body image cluster; including: randomly selecting a maternal body image cluster as a reference maternal body image cluster; and selecting the maternal body image with the largest reference coefficient in the reference maternal body image cluster as the target image; The difference between the target image and each other part-time body image in the reference part-time body image cluster is calculated to obtain the residual image corresponding to each other part-time body image in the reference part-time body image cluster; all the residual images are quantized and encoded to obtain the encoding result of the reference part-time body image cluster. Traverse each maternal body image cluster to obtain the encoding result of each maternal body image cluster; The method for obtaining the motion feature values ​​includes: The motion feature value is obtained according to the motion feature value calculation formula, which is as follows: ； In the formula, Represents the motion feature values ​​of the reference frame image; Indicates the number of edges in the reference frame image; This indicates the number of edges in the previous frame of the mother's body image in the reference frame image; This indicates the number of edges in the next frame of the mother's body image following the reference frame image. Indicates the reference frame image with the first The preset number of other edges that are closest to each edge; Indicates the reference frame image with the first The nearest edge distance is the first The curvature of the other edges; Indicates the reference frame image with the first The nearest edge distance is the first The number of pixels in each of the other edges; Indicates the reference frame image with the first The nearest edge distance is the first The distance between the endpoints of the other edges; Indicates the first The edge and the nearest one The distance between the other edges; Represents the absolute value function; The method for obtaining the reference coefficient includes: Calculate the information entropy of the grayscale values ​​of all other pixels within a preset neighborhood of each pixel in each frame of the maternal body image, and use this as the reference information entropy for each pixel. Sum the reference information entropies of all pixels forming the edges in the reference maternal body image cluster to obtain the reference coefficient for each frame of the maternal body image. The formula for calculating the reference coefficient is as follows: ； In the formula, This represents the reference coefficient for each frame of the mother's body image; This indicates the number of edges in each frame of the mother's body image; Indicates the first The number of pixels contained in each edge; Indicates the first The edge of the first The information entropy of the grayscale values ​​of pixels within a preset neighborhood of a given pixel is set to the value of the pixel at the first pixel. Centered on 100 pixels A rectangular area.

2. The method for remote monitoring and management of postpartum recovery of mothers according to claim 1, characterized in that, The method for obtaining the motion coefficients includes: The difference region between two adjacent frames of the mother's body image at the reference time is obtained using the frame difference method. The motion coefficient is obtained according to the motion coefficient calculation formula, which is shown below: ； In the formula, Indicates the motion coefficients of the image at the reference time. This indicates the number of pixels in the difference region between the previous frame of the mother's body image and the reference time image. This represents the number of pixels in the difference region between the next frame of the mother's body image at the reference time and the reference time image. This represents the information entropy in the image at the reference time. The information entropy of the image of the mother's body in the previous frame of the reference time image; The entropy represents the information of the next frame of the mother's body image at the reference time. This represents the maximum value function.

3. The method for remote monitoring and management of postpartum recovery of mothers according to claim 1, characterized in that, The method for obtaining the dynamic-static interval frame image includes: The first frame of the mother's image with a motion coefficient greater than a preset first threshold is used as the static-dynamic interval frame image.

4. The method for remote monitoring and management of postpartum recovery of mothers according to claim 1, characterized in that, The method for acquiring the maternal body image cluster includes: Calculate the mean motion feature value of all frames of maternal body images in each pair of adjacent dynamic and static interval frames to obtain the first mean; use the first mean to cluster all maternal body images into dynamic maternal image clusters and static maternal image clusters. The dynamic and static maternal image clusters are clustered twice using the motion feature values ​​of each frame of maternal body image to obtain all maternal body image clusters.

5. A remote monitoring and management system for postpartum recovery, the system comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that, When the processor executes the computer program, it implements the steps of the remote monitoring and management method for postpartum recovery of mothers as described in any one of claims 1 to 4.

6. A computer-readable storage medium storing a computer program, characterized in that, When the computer program is executed by the processor, it implements the steps of the remote monitoring and management method for postpartum recovery of mothers as described in any one of claims 1 to 4.

7. A computer device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that, When the processor executes the computer program, it implements the steps of the remote monitoring and management method for postpartum recovery of mothers as described in any one of claims 1 to 4.

Images