CN114120177A

CN114120177A - User behavior guiding method and system based on intelligent health monitoring

Info

Publication number: CN114120177A
Application number: CN202111333577.0A
Authority: CN
Inventors: 邹雪; 聂炎; 张涛; 晏飞
Original assignee: Individual
Current assignee: Individual
Priority date: 2021-11-11
Filing date: 2021-11-11
Publication date: 2022-03-01

Abstract

The invention provides a user behavior guiding method and system based on intelligent health monitoring, and relates to the technical field of health monitoring. In the invention, a plurality of acquired monitoring videos of medical places are processed to obtain a plurality of corresponding target monitoring videos; for each target monitoring video, carrying out user analysis processing on the target monitoring video to obtain a corresponding user analysis result, and processing the target monitoring video based on the user analysis result to obtain at least one corresponding monitoring video to be identified, wherein the user objects in each two frames of monitoring video frames of the medical place in one monitoring video to be identified are the same; and aiming at each monitoring video to be identified, carrying out user behavior guide processing on a user object in the monitoring video to be identified based on the monitoring video frame of the medical place, which is included in the monitoring video to be identified. Based on the method, the problem of poor effect on user behavior guidance in the prior art can be solved.

Description

User behavior guiding method and system based on intelligent health monitoring

Technical Field

The invention relates to the technical field of health monitoring, in particular to a user behavior guiding method and system based on intelligent health monitoring.

Background

With the coming of the network era, the depth, the breadth and the mode of connection among organizations such as people, enterprises, governments and the like, intelligent robots and intelligent articles are continuously expanded by technologies such as big data, artificial intelligence, internet of things, industrial 4.0, cloud computing and the like. The group phenomena under the network environment are connected more closely, and a crowd-sourcing network is formed. The concept of the crowd-sourcing science aims to explore the basic principle and the law of group intelligent activities formed by three systems of information, physics and society under the background of large-scale online interconnection.

With the rapid development of intelligent health, the requirement for medical informatization has moved from the simple application of collected data to the data utilization stage. The intelligent analysis and big data mining technology is used for solving the problems of medical treatment and treatment of patients, and the problem is a necessary trend for the development of the future medical health industry. More and more patients choose to acquire information related to medical services in an information-based manner, such as hospital information, doctor expertise, and drug services available in drug stores. In which, making appointment, purchasing medicine and feeding back illness state through on-line mode are becoming mainstream interactive mode between patients and intelligent agents such as hospitals, doctors and drugstores. In addition, in some application scenarios, a medical place is complicated, and it is difficult for a user to perform corresponding activities accurately, for this reason, in the prior art, a corresponding visit guide staff is generally configured, so that the user can make an inquiry, but the user does not actually make an inquiry actively, which may result in a problem of poor effect of guiding the user behavior.

Disclosure of Invention

In view of the above, the present invention provides a method and a system for guiding user behavior based on intelligent health monitoring to solve the problem of poor effect of guiding user behavior in the prior art.

In order to achieve the above purpose, the embodiment of the invention adopts the following technical scheme:

a user behavior guiding method based on intelligent health monitoring is applied to a medical monitoring server, the medical monitoring server is in communication connection with a plurality of medical monitoring terminal devices, and the user behavior guiding method based on intelligent health monitoring comprises the following steps:

processing a plurality of acquired medical site monitoring videos acquired by the plurality of medical monitoring terminal devices to obtain a plurality of corresponding target monitoring videos, wherein each medical site monitoring video comprises a plurality of medical site monitoring video frames, each medical site monitoring video frame is obtained by acquiring an image of a corresponding monitoring site area based on the corresponding medical monitoring terminal device, and each target monitoring video comprises at least one medical site monitoring video frame;

for each target monitoring video in the target monitoring videos, performing user analysis processing on the target monitoring video to obtain a user analysis result corresponding to the target monitoring video, and processing the target monitoring video based on the user analysis result to obtain at least one to-be-identified monitoring video corresponding to the target monitoring video, wherein for each to-be-identified monitoring video, user objects in every two frames of monitoring video frames of the medical place are the same;

and aiming at each monitoring video to be identified in the at least one monitoring video to be identified, carrying out user behavior guidance processing on a user object in the monitoring video to be identified based on the monitoring video frame of the medical place, which is included in the monitoring video to be identified.

In some preferred embodiments, in the method for guiding user behavior based on intelligent health monitoring, the step of performing user analysis processing on each target surveillance video of the plurality of target surveillance videos to obtain a user analysis result corresponding to the target surveillance video, and processing the target surveillance video based on the user analysis result to obtain at least one to-be-identified surveillance video corresponding to the target surveillance video includes:

for each target monitoring video in the target monitoring videos, performing object identification processing on each frame of monitoring video frame of the medical place included in the target monitoring video to determine each user object corresponding to the target monitoring video;

counting the number of user objects corresponding to the monitoring video frames of the medical places, which are included in the target monitoring video, aiming at each target monitoring video in the target monitoring videos to obtain the number of video frame objects corresponding to the target monitoring video;

for each target surveillance video in the multiple target surveillance videos, determining a relative size relationship between the number of video frame objects corresponding to the target surveillance video and a preset object number threshold, and determining the target surveillance video as a first target surveillance video when the number of video frame objects is less than the object number threshold, or determining the target surveillance video as a second target surveillance video when the number of video frame objects is greater than or equal to the object number threshold;

determining, for each of the first target surveillance videos, whether a user object in each of two frames of medical site surveillance video frames included in the first target surveillance video is the same when at least one of the first target surveillance videos exists, and when the user objects in every two frames of monitoring video frames of the medical place included in the first target monitoring video are the same, determining the first target monitoring video as the monitoring video to be identified, and screening out each medical site surveillance video frame having a user object when the user objects in at least two medical site surveillance video frames included in the first target surveillance video are different, classifying the monitoring video frames of the medical place based on whether the user objects are the same or not to obtain at least one corresponding video frame set, and respectively taking each video frame set as a monitoring video to be identified corresponding to a user object.

In some preferred embodiments, in the method for guiding user behavior based on intelligent health monitoring, the step of performing user analysis processing on each target surveillance video of the plurality of target surveillance videos to obtain a user analysis result corresponding to the target surveillance video, and processing the target surveillance video based on the user analysis result to obtain at least one to-be-identified surveillance video corresponding to the target surveillance video further includes:

when at least one second target monitoring video exists, decomposing each medical place monitoring video frame in each second target monitoring video based on the number of user objects of the medical place monitoring video frame to obtain a new medical place monitoring video frame of a corresponding number of frames for replacing the medical place monitoring video frame;

and clustering the medical place monitoring video frames and/or the new medical place monitoring video frames included in the second target monitoring video based on whether the corresponding user objects are the same or not for each second target monitoring video to obtain at least one corresponding video frame clustering set, and respectively clustering each video frame into a monitoring video to be identified corresponding to one user object.

In some preferred embodiments, in the method for guiding user behavior based on intelligent health monitoring, the clustering, for each second target surveillance video, the medical-site surveillance video frames and/or the new medical-site surveillance video frames included in the second target surveillance video based on whether the corresponding user object is the same or not to obtain at least one corresponding video-frame cluster set, and respectively clustering each video-frame cluster set as a surveillance video to be identified corresponding to one user object includes:

clustering the medical place monitoring video frames and/or the new medical place monitoring video frames included in the second target monitoring video based on whether the corresponding user objects are the same or not for each second target monitoring video to obtain at least one corresponding video frame clustering set;

and determining the monitored video to be identified corresponding to each user object according to each two video frame cluster sets in the at least one video frame cluster set and based on whether the user objects corresponding to the two video frame cluster sets are the same or not.

In some preferred embodiments, in the method for guiding user behavior based on intelligent health monitoring, the step of determining, for every two video frame cluster sets in the at least one video frame cluster set, a monitored video to be identified corresponding to each user object based on whether user objects corresponding to the two video frame cluster sets are the same or not includes:

determining whether user objects corresponding to the two video frame cluster sets are the same or not aiming at every two video frame cluster sets in the at least one video frame cluster set;

for every two video frame cluster sets in the at least one video frame cluster set, if the user objects corresponding to the two video frame cluster sets are different, determining the two video frame cluster sets as two corresponding monitored videos to be identified respectively;

and aiming at every two video frame cluster sets in the at least one video frame cluster set, if the user objects corresponding to the two video frame cluster sets are the same, merging the two video frame cluster sets into a corresponding monitoring video to be identified.

In some preferred embodiments, in the above method for guiding user behavior based on intelligent health monitoring, the step of, for each monitored video to be identified in the at least one monitored video to be identified, performing user behavior guidance processing on a user object in the monitored video to be identified based on the medical site monitored video frame included in the monitored video to be identified includes:

performing action recognition processing on the surveillance video to be recognized aiming at each surveillance video to be recognized in the at least one surveillance video to be recognized to obtain an action recognition result corresponding to the surveillance video to be recognized, and determining whether user behavior guidance processing needs to be performed on a user object in the surveillance video to be recognized or not based on the action recognition result;

and aiming at each monitored video to be identified in the at least one monitored video to be identified, if the user behavior guidance processing needs to be carried out on the user object in the monitored video to be identified, carrying out the user behavior guidance processing on the user object in the monitored video to be identified.

In some preferred embodiments, in the above method for guiding user behavior based on intelligent health monitoring, the step of performing, for each to-be-identified surveillance video in the at least one to-be-identified surveillance video, motion recognition processing on the to-be-identified surveillance video to obtain a motion recognition result corresponding to the to-be-identified surveillance video, and determining whether user behavior guidance processing needs to be performed on a user object in the to-be-identified surveillance video based on the motion recognition result includes:

performing action recognition processing on the surveillance video to be recognized aiming at each surveillance video to be recognized in the at least one surveillance video to be recognized to obtain an action recognition result corresponding to the surveillance video to be recognized, and determining whether the user object corresponding to the surveillance video to be recognized has repeated same actions or not based on the action recognition result;

for each monitored video to be identified in the at least one monitored video to be identified, if the user object corresponding to the monitored video to be identified has the same action repeated for multiple times, it is determined that user behavior guidance processing needs to be performed on the user object in the monitored video to be identified, and if the user object corresponding to the monitored video to be identified does not have the same action repeated for multiple times, it is determined that user behavior guidance processing does not need to be performed on the user object in the monitored video to be identified.

The embodiment of the invention also provides a user behavior guidance system based on intelligent health monitoring, which is applied to a medical monitoring server, wherein the medical monitoring server is in communication connection with a plurality of medical monitoring terminal devices, and the user behavior guidance system based on intelligent health monitoring comprises:

the first video processing module is used for processing a plurality of acquired medical site monitoring videos acquired by the plurality of medical monitoring terminal devices to obtain a plurality of corresponding target monitoring videos, wherein each medical site monitoring video comprises a plurality of medical site monitoring video frames, each medical site monitoring video frame is obtained by acquiring an image of a corresponding monitoring site area based on the corresponding medical monitoring terminal device, and each target monitoring video comprises at least one medical site monitoring video frame;

the second video processing module is used for carrying out user analysis processing on the target monitoring video aiming at each target monitoring video in the target monitoring videos to obtain a user analysis result corresponding to the target monitoring video and processing the target monitoring video based on the user analysis result to obtain at least one to-be-identified monitoring video corresponding to the target monitoring video, wherein for each to-be-identified monitoring video, user objects in every two frames of monitoring video frames of the medical place are the same;

and the user behavior guide module is used for carrying out user behavior guide processing on a user object in the monitoring video to be identified based on the monitoring video frame of the medical place, wherein the monitoring video to be identified comprises each monitoring video to be identified in the at least one monitoring video to be identified.

In some preferred embodiments, in the above system for guiding user behavior based on intelligent health monitoring, the second video processing module is specifically configured to:

In some preferred embodiments, in the above system for guiding user behavior based on intelligent health monitoring, the user behavior guiding module is specifically configured to:

The embodiment of the invention provides a user behavior guiding method and a system based on intelligent health monitoring, after the obtained multiple medical site monitoring videos collected by the multiple medical monitoring terminal devices are processed to obtain corresponding multiple target monitoring videos, each target monitoring video can be firstly targeted, performing user analysis processing on the target monitoring video to obtain a user analysis result corresponding to the target monitoring video, and processing the target monitoring video based on the user analysis result to obtain at least one to-be-identified monitoring video corresponding to the target monitoring video, so that the user behavior guiding processing can be carried out on the user object corresponding to each monitoring video to be identified based on the monitoring video to be identified, namely, the user behavior guidance processing is actively carried out on the user object, so that the problem of poor effect on user behavior guidance in the prior art is solved.

In order to make the aforementioned and other objects, features and advantages of the present invention comprehensible, preferred embodiments accompanied with figures are described in detail below.

Drawings

Fig. 1 is an application block diagram of a medical monitoring server according to an embodiment of the present invention.

Fig. 2 is a schematic diagram of a method for guiding user behavior based on intelligent health monitoring according to an embodiment of the present invention.

Fig. 3 is a schematic diagram of a system for guiding user behavior based on intelligent health monitoring according to an embodiment of the present invention.

Detailed Description

In order to make the objects, technical solutions and advantages of the embodiments of the present invention clearer, the technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the drawings in the embodiments of the present invention, and it is obvious that the described embodiments are only a part of the embodiments of the present invention, and not all of the embodiments. The components of embodiments of the present invention generally described and illustrated in the figures herein may be arranged and designed in a wide variety of different configurations.

Thus, the following detailed description of the embodiments of the present invention, presented in the figures, is not intended to limit the scope of the invention, as claimed, but is merely representative of selected embodiments of the invention. All other embodiments, which can be derived by a person skilled in the art from the embodiments given herein without making any creative effort, shall fall within the protection scope of the present invention.

As shown in fig. 1, an embodiment of the present invention provides a medical monitoring server. Wherein the medical monitoring server may include a memory and a processor. The memory and the processor are electrically connected directly or indirectly to enable data transfer or interaction. For example, they may be electrically connected to each other via one or more communication buses or signal lines. The memory can have stored therein at least one software function (computer program) which can be present in the form of software or firmware. The processor may be configured to execute the executable computer program stored in the memory to implement the method for guiding user behavior based on intelligent health monitoring provided by the embodiment of the present invention.

For example, in some preferred embodiments, the Memory may be, but is not limited to, Random Access Memory (RAM), Read Only Memory (ROM), Programmable Read-Only Memory (PROM), Erasable Read-Only Memory (EPROM), electrically Erasable Read-Only Memory (EEPROM), and the like. The Processor may be a general-purpose Processor, including a Central Processing Unit (CPU), a Network Processor (NP), a System on Chip (SoC), and the like; but may also be a Digital Signal Processor (DSP), an Application Specific Integrated Circuit (ASIC), a Field Programmable Gate Array (FPGA) or other programmable logic device, discrete gate or transistor logic device, discrete hardware components.

For example, in some preferred embodiments, the structure shown in fig. 1 is merely illustrative, and the medical monitoring server may further include more or fewer components than those shown in fig. 1, or have a different configuration than that shown in fig. 1, for example, may include a communication unit for information interaction with other devices.

With reference to fig. 2, an embodiment of the present invention further provides a user behavior guiding method based on intelligent health monitoring, which can be applied to the medical monitoring server. The method steps defined by the flow related to the intelligent health monitoring-based user behavior guidance method can be realized by the medical monitoring server, and the medical monitoring server is in communication connection with a plurality of medical monitoring terminal devices.

The specific process shown in FIG. 2 will be described in detail below.

And S100, processing the acquired multiple medical site monitoring videos acquired by the multiple medical monitoring terminal devices to obtain corresponding multiple target monitoring videos.

In the embodiment of the present invention, the medical monitoring server may process the acquired multiple medical site monitoring videos collected by the multiple medical monitoring terminal devices to obtain multiple corresponding target monitoring videos, that is, execute step S100. Each medical site monitoring video comprises a plurality of frames of medical site monitoring video frames, each frame of medical site monitoring video frame is obtained by carrying out image acquisition on a corresponding monitoring site area based on the corresponding medical monitoring terminal equipment, and each target monitoring video comprises at least one frame of medical site monitoring video frame.

Step S200, aiming at each target monitoring video in the target monitoring videos, carrying out user analysis processing on the target monitoring video to obtain a user analysis result corresponding to the target monitoring video, and processing the target monitoring video based on the user analysis result to obtain at least one monitoring video to be identified corresponding to the target monitoring video.

In the embodiment of the present invention, the medical monitoring server may perform user parsing on each target monitoring video of the plurality of target monitoring videos to obtain a user parsing result corresponding to the target monitoring video, and process the target monitoring video based on the user parsing result to obtain at least one to-be-identified monitoring video corresponding to the target monitoring video, that is, perform step S200. And for each monitoring video to be identified, the user objects in every two frames of monitoring video frames of the medical site included in the monitoring video to be identified are the same.

Step S300, aiming at each monitoring video to be identified in the at least one monitoring video to be identified, carrying out user behavior guiding processing on a user object in the monitoring video to be identified based on the monitoring video frame of the medical place, wherein the monitoring video to be identified comprises the monitoring video frame of the medical place.

In this embodiment of the present invention, the medical monitoring server may perform, for each to-be-identified monitoring video in the at least one to-be-identified monitoring video, user behavior guidance processing on a user object in the to-be-identified monitoring video based on the medical site monitoring video frame included in the to-be-identified monitoring video, that is, execute step S300.

Based on the user behavior guidance method, after the obtained multiple medical site surveillance videos collected by the multiple medical monitoring terminal devices are processed to obtain the corresponding multiple target surveillance videos, user analysis processing can be performed on the target surveillance videos to obtain user analysis results corresponding to the target surveillance videos, and the target surveillance videos are processed based on the user analysis results to obtain at least one to-be-identified surveillance video corresponding to the target surveillance videos, so that user behavior guidance processing can be performed on the corresponding user object based on the to-be-identified surveillance videos for each to-be-identified surveillance video, that is, user behavior guidance processing can be actively performed on the user object, and therefore the problem that the effect of user behavior guidance in the prior art is poor is solved.

For example, in some preferred embodiments, the step of processing the acquired multiple medical site monitoring videos collected by the multiple medical monitoring terminal devices to obtain corresponding multiple target monitoring videos, that is, step S100, may include the following steps S110, S120, and S130.

Step S110, obtaining medical site monitoring videos respectively collected by the plurality of medical monitoring terminal devices, and obtaining a plurality of medical site monitoring videos corresponding to the plurality of medical monitoring terminal devices.

In the embodiment of the present invention, the medical monitoring server may obtain the medical site monitoring videos respectively acquired by the plurality of medical monitoring terminal devices, to obtain a plurality of medical site monitoring videos corresponding to the plurality of medical monitoring terminal devices, that is, to execute step S110. Each medical place monitoring video in the plurality of medical place monitoring videos comprises a plurality of medical place monitoring video frames, and each medical place monitoring video frame is obtained by carrying out image acquisition on a corresponding monitoring place area based on the corresponding medical monitoring terminal equipment.

Step S120, analyzing each of the plurality of medical site surveillance videos respectively to determine whether each of the plurality of medical site surveillance videos needs to be screened.

In this embodiment of the present invention, the medical monitoring server may analyze each of the plurality of medical site monitoring videos, respectively, to determine whether each of the plurality of medical site monitoring videos needs to be filtered, that is, step S120 is executed.

Step S130, for each of the plurality of medical site surveillance videos, if it is determined that the medical site surveillance video needs to be screened, screening the medical site surveillance video to obtain a target surveillance video corresponding to the medical site surveillance video.

In the embodiment of the present invention, the medical monitoring server may perform, for each of the plurality of medical site monitoring videos, if it is determined that the medical site monitoring video needs to be screened, screening the medical site monitoring video to obtain a target monitoring video corresponding to the medical site monitoring video, that is, perform step S130. Wherein each target monitoring video comprises at least one frame of the medical site monitoring video.

Therefore, after a plurality of medical place monitoring videos corresponding to a plurality of medical monitoring terminal devices are obtained, each of the plurality of medical place monitoring videos can be analyzed respectively to determine whether each of the plurality of medical place monitoring videos needs to be screened, then the medical place monitoring videos needing to be screened can be screened to obtain corresponding target monitoring videos, and therefore screening of the medical place monitoring videos can be achieved through an effective judgment mechanism, effective management of the medical place monitoring videos is achieved, screening is conducted when necessary, and the problem that management effects of the medical place monitoring videos in the prior art are poor is solved.

For example, in some preferred embodiments, the step of obtaining a plurality of medical site monitoring videos corresponding to the plurality of medical monitoring terminal devices by obtaining the medical site monitoring videos respectively collected by the plurality of medical monitoring terminal devices, that is, step S110, may include the following steps:

firstly, judging whether medical place monitoring videos respectively acquired by the medical monitoring terminal devices can be acquired or not, and generating first video acquisition notification information when the medical place monitoring videos respectively acquired by the medical monitoring terminal devices are judged to be acquired;

secondly, sending the first video acquisition notification information to each medical monitoring terminal device in the plurality of medical monitoring terminal devices, wherein each medical monitoring terminal device is used for sending the currently acquired monitoring video of the medical place to the medical monitoring server after receiving the first video acquisition notification information;

then, each medical monitoring terminal device in the plurality of medical monitoring terminal devices acquires the medical place monitoring video sent by the notification information based on the first video, and a plurality of medical place monitoring videos corresponding to the plurality of medical monitoring terminal devices are obtained.

For example, in some preferred embodiments, the step of determining whether the medical site monitoring videos collected by the plurality of medical monitoring terminal devices can be acquired respectively, and generating the first video acquisition notification information when it is determined that the medical site monitoring videos collected by the plurality of medical monitoring terminal devices can be acquired respectively may include the following steps:

firstly, judging whether video sending confirmation information sent by each medical monitoring terminal device in the plurality of medical monitoring terminal devices is obtained or not, and judging that medical site monitoring videos respectively collected by the plurality of medical monitoring terminal devices can be obtained after the video sending confirmation information sent by each medical monitoring terminal device in the plurality of medical monitoring terminal devices is obtained;

and secondly, when the medical site monitoring videos respectively acquired by the plurality of medical monitoring terminal devices can be acquired, generating first video acquisition notification information.

For another example, in another preferred embodiment, the step of determining whether the medical site monitoring videos collected by the plurality of medical monitoring terminal devices can be acquired respectively, and generating the first video acquisition notification information when it is determined that the medical site monitoring videos collected by the plurality of medical monitoring terminal devices can be acquired respectively may include the following steps:

firstly, calculating target time-consuming duration for the medical monitoring server to process and complete a current data calculation task, and determining the relative size between the target time-consuming duration and a preset time-consuming duration threshold;

secondly, when the target time-consuming duration is greater than the time-consuming duration threshold, judging that the medical site monitoring videos respectively acquired by the medical monitoring terminal devices cannot be acquired;

then, when the target time-consuming duration is less than or equal to the time-consuming duration threshold, it is determined that the medical site monitoring videos respectively acquired by the plurality of medical monitoring terminal devices can be acquired, and corresponding first video acquisition notification information is generated.

For example, on the basis of the above example, in some preferred embodiments, the step of determining whether the medical site monitoring videos collected by the plurality of medical monitoring terminal devices respectively can be acquired, and generating the first video acquisition notification information when it is determined that the medical site monitoring videos collected by the plurality of medical monitoring terminal devices respectively can be acquired, may further include the following steps:

firstly, when the target time-consuming duration is greater than the time-consuming duration threshold, generating first video removal notification information, and sending the first video removal notification information to each of the plurality of medical monitoring terminal devices, wherein each of the medical monitoring terminal devices is configured to delete a currently acquired medical site monitoring video after receiving the first video removal notification information, and continue to acquire a new medical site monitoring video.

For example, in some preferred embodiments, the step of parsing each of the plurality of medical site monitoring videos to determine whether each of the medical site monitoring videos needs to be filtered, that is, step S120, may include the following steps:

firstly, aiming at each medical place monitoring video in the plurality of medical place monitoring videos, carrying out object identification processing on a plurality of medical place monitoring video frames included in the medical place monitoring video to obtain an object identification result corresponding to the medical place monitoring video;

secondly, whether each medical place monitoring video needs to be screened or not is determined based on the object identification result corresponding to each medical place monitoring video in the plurality of medical place monitoring videos.

For example, in some preferred embodiments, the step of determining whether each of the plurality of medical site monitoring videos needs to be filtered based on the object identification result corresponding to each of the plurality of medical site monitoring videos may include the following steps:

firstly, for each medical place monitoring video in the plurality of medical place monitoring videos, determining the number of user objects in a plurality of medical place monitoring video frames included in the medical place monitoring video based on an object identification result corresponding to the medical place monitoring video, and obtaining the statistical number of the user objects corresponding to the medical place monitoring video, wherein the user objects include patients and accompanying persons (all persons appearing in the medical place monitoring video);

secondly, counting the sum of the user object counting numbers respectively corresponding to the plurality of medical place monitoring videos to obtain the user object counting total number corresponding to the plurality of medical place monitoring videos;

then, whether each medical site monitoring video needs to be screened is determined based on the total number of the user object statistics corresponding to the plurality of medical site monitoring videos (if the total number of the user object statistics is greater than or equal to a threshold, it may be determined that screening is not needed).

For another example, in another preferred embodiment, the step of determining whether each medical site monitoring video needs to be filtered based on the object identification result corresponding to each medical site monitoring video in the plurality of medical site monitoring videos may also include the following steps:

secondly, determining the area of a monitoring place region corresponding to the medical monitoring terminal equipment corresponding to the medical place monitoring video aiming at each medical place monitoring video in the plurality of medical place monitoring videos, and calculating to obtain the distribution density information of the user object corresponding to the medical place monitoring video based on the area of the region and the statistical number of the user object corresponding to the medical place monitoring video;

then, acquiring a medical place monitoring audio acquired by an audio monitoring terminal device deployed in a monitoring place area corresponding to the medical monitoring terminal device corresponding to the medical place monitoring video for each of the plurality of medical place monitoring videos, and performing sound source identification processing on the medical place monitoring audio to obtain the number of sound sources corresponding to the medical place monitoring audio;

then, for each medical place monitoring video in the multiple medical place monitoring videos, calculating to obtain user sound source distribution density information corresponding to the medical place monitoring video based on the number of sound sources corresponding to the medical place monitoring audio corresponding to the medical place monitoring video and the area of the corresponding monitoring place area, and fusing (for example, calculating an average value) the user sound source distribution density information and the user object distribution density information corresponding to the medical place monitoring video to obtain target distribution density information corresponding to the medical place monitoring video, wherein the target distribution density information is used for representing the user complexity degree in the corresponding monitoring place area;

and finally, determining whether each medical place monitoring video needs to be screened or not based on the target distribution density information respectively corresponding to the plurality of medical place monitoring videos.

For example, in some preferred embodiments, the step of determining whether each medical site monitoring video needs to be filtered based on the target distribution density information corresponding to each of the plurality of medical site monitoring videos may include the following steps:

firstly, calculating an average value of the target distribution density information corresponding to the plurality of medical site monitoring videos respectively to obtain distribution density average value information corresponding to the plurality of medical site monitoring videos, and determining the size between the distribution density average value information and pre-configured distribution density threshold value information (for example, the distribution density average value information is greater than or equal to the distribution density threshold value);

secondly, if the distribution density mean value information is larger than or equal to the distribution density threshold, determining that each medical site monitoring video does not need to be screened, and if the distribution density mean value information is smaller than the distribution density threshold, determining the size relationship between the target distribution density information corresponding to the medical site monitoring video and the distribution density threshold for each medical site monitoring video in the plurality of medical site monitoring videos;

then, for each of the plurality of medical site surveillance videos, if the target distribution density information corresponding to the medical site surveillance video is greater than or equal to the distribution density threshold, it is determined that the medical site surveillance video does not need to be screened, and if the target distribution density information corresponding to the medical site surveillance video is less than the distribution density threshold, it is determined that the medical site surveillance video needs to be screened.

For example, in some preferred embodiments, if it is determined that the medical site surveillance video needs to be filtered for each of the multiple medical site surveillance videos, the step of filtering the medical site surveillance video to obtain a target surveillance video corresponding to the medical site surveillance video, that is, step S130, may include the following steps:

firstly, aiming at each of the plurality of medical place monitoring videos, if it is determined that the medical place monitoring video needs to be screened, determining a target screening proportion (for example, a negative correlation relationship may exist between the target distribution density information and the target screening proportion) for screening the medical place monitoring video based on the target distribution density information corresponding to the medical place monitoring video, wherein the target screening proportion is used for representing a maximum ratio of medical place monitoring video frames screened after the corresponding medical place monitoring video is screened;

secondly, aiming at each of the plurality of medical place monitoring videos, if it is determined that the medical place monitoring video needs to be screened, performing target screening operation on a plurality of medical place monitoring video frames included in the medical place monitoring video based on the target screening proportion corresponding to the medical place monitoring video so as to obtain a target monitoring view corresponding to the medical place monitoring video.

For example, in some preferred embodiments, for each of the plurality of medical site monitoring videos, if it is determined that the medical site monitoring video needs to be screened, performing a target screening operation on multiple frames of medical site monitoring video frames included in the medical site monitoring video based on the target screening ratio corresponding to the medical site monitoring video to obtain a target monitoring video corresponding to the medical site monitoring video, where the target screening operation in this step further includes the following steps:

firstly, performing video frame time sequence exchange processing (random exchange can be performed) on at least two frames of medical place monitoring video frames included in the medical place monitoring video to obtain an updated medical place monitoring video corresponding to the medical place monitoring video;

secondly, determining at least one video frame replacement combination based on the medical site monitoring video and the updated medical site monitoring video, wherein one video frame replacement combination corresponds to one processing time sequence exchange in the video frame time sequence exchange processing, and each video frame replacement combination comprises one medical site monitoring video frame corresponding to one time sequence exchange in the medical site monitoring video and one medical site monitoring video frame corresponding to the time sequence exchange in the updated medical site monitoring video (namely two medical site monitoring video frames before and after the exchange);

then, aiming at each video frame replacement combination in the at least one video frame replacement combination, determining whether the video frame similarity between two monitoring video frames of the medical field in the video frame replacement combination is greater than or equal to a first video frame similarity threshold value configured in advance;

then, aiming at each video frame replacement combination in the at least one video frame replacement combination, if the video frame similarity between two medical site monitoring video frames in the video frame replacement combination is greater than or equal to the first video frame similarity threshold, determining the two medical site monitoring video frames in the video frame replacement combination as the medical site monitoring video frames to be screened;

and finally, respectively calculating the video frame similarity between every two frames of the monitoring video frames of the medical places to be screened, and performing duplicate removal screening processing on at least two obtained frames of the monitoring video frames of the medical places to be screened on the basis of the video frame similarity between every two frames of the monitoring video frames of the medical places to be screened and the target screening proportion corresponding to the monitoring video of the medical places to obtain the target monitoring video corresponding to the monitoring video of the medical places.

For example, in some preferred embodiments, the step of determining at least one alternative combination of video frames based on the healthcare facility monitoring video and the updated healthcare facility monitoring video may further include the steps of:

firstly, performing video frame comparison processing on the medical place monitoring video and the updated medical place monitoring video based on a first preset granularity to obtain a corresponding first video frame comparison result sequence, wherein the first video frame comparison result sequence comprises at least one first video frame combination, and each first video frame combination is used for identifying a corresponding pair of medical place monitoring video frames between the medical place monitoring video and the updated medical place monitoring video and the video frame similarity corresponding to the pair of medical place monitoring video frames;

secondly, decomposing a first video frame combination in the first video frame comparison result sequence based on a second preset granularity to obtain a second video frame comparison result sequence corresponding to the first video frame comparison result sequence, wherein the second video frame comparison result sequence comprises at least one second video frame combination, each second video frame combination is used for identifying a pair of corresponding medical place monitoring video frames between the medical place monitoring video and the updated medical place monitoring video and the similarity of the video frames corresponding to the pair of medical place monitoring video frames, and the second granularity is smaller than the first granularity;

then, based on the video frame similarity corresponding to each second video frame combination and each second video frame combination in at least one second video frame combination included in the second video frame comparison result sequence, determining at least one video frame replacement combination to perform video frame comparison processing on the medical site surveillance video and the updated medical site surveillance video based on a first preset granularity to obtain a corresponding first video frame comparison result sequence, wherein the first video frame comparison result sequence includes at least one first video frame combination, each first video frame combination is used for identifying a corresponding pair of medical site surveillance video frames between the medical site surveillance video and the updated medical site surveillance video, and the video frame similarity corresponding to the pair of medical site surveillance video frames;

then, decomposing a first video frame combination in the first video frame comparison result sequence based on a second preset granularity to obtain a second video frame comparison result sequence corresponding to the first video frame comparison result sequence, wherein the second video frame comparison result sequence comprises at least one second video frame combination, each second video frame combination is used for identifying a pair of corresponding medical place monitoring video frames between the medical place monitoring video and the updated medical place monitoring video and the video frame similarity corresponding to the pair of medical place monitoring video frames, and the second granularity is smaller than the first granularity;

and finally, determining at least one video frame replacement combination based on the video frame similarity corresponding to each of the at least one second video frame combination and each second video frame combination included in the second video frame comparison result sequence (if the video frame similarity corresponding to the second video frame combination is smaller than a preset second video frame similarity threshold value).

In addition, on the basis of the above embodiment, for each of the plurality of medical site monitoring videos, if it is determined that the medical site monitoring video does not need to be screened, the medical site monitoring video is used as the target monitoring video corresponding to the medical site monitoring video.

For example, in some preferred embodiments, the step of, for each target surveillance video of the target surveillance videos, performing user analysis processing on the target surveillance video to obtain a user analysis result corresponding to the target surveillance video, and processing the target surveillance video based on the user analysis result to obtain at least one surveillance video to be identified corresponding to the target surveillance video, that is, step S200, may include the following steps:

firstly, aiming at each target monitoring video in the target monitoring videos, carrying out object recognition processing on each frame of monitoring video frame of a medical place included in the target monitoring video, and determining each user object corresponding to the target monitoring video;

secondly, counting the number of user objects corresponding to the monitoring video frames of the medical places, which are included in the target monitoring video, aiming at each target monitoring video in the plurality of target monitoring videos to obtain the number of video frame objects corresponding to the target monitoring video;

then, for each target surveillance video in the multiple target surveillance videos, determining a relative size relationship between the number of video frame objects corresponding to the target surveillance video and a preset object number threshold, and determining the target surveillance video as a first target surveillance video when the number of video frame objects is smaller than the object number threshold, or determining the target surveillance video as a second target surveillance video when the number of video frame objects is larger than or equal to the object number threshold;

finally, when at least one first target monitoring video exists, determining whether the user objects in every two frames of medical site monitoring video frames included in the first target monitoring video are the same or not for each first target monitoring video, and when the user objects in every two frames of monitoring video frames of the medical place included in the first target monitoring video are the same, determining the first target monitoring video as the monitoring video to be identified, and screening out each medical site surveillance video frame having a user object when the user objects in at least two medical site surveillance video frames included in the first target surveillance video are different, classifying the monitoring video frames of the medical place based on whether the user objects are the same or not to obtain at least one corresponding video frame set, and respectively taking each video frame set as a monitoring video to be identified corresponding to a user object.

For example, in some preferred embodiments, the step of, for each target surveillance video of the target surveillance videos, performing user analysis processing on the target surveillance video to obtain a user analysis result corresponding to the target surveillance video, and processing the target surveillance video based on the user analysis result to obtain at least one surveillance video to be identified corresponding to the target surveillance video, that is, step S200, may further include the following steps:

firstly, when at least one second target monitoring video exists, decomposing each medical place monitoring video frame in each second target monitoring video based on the number of user objects in the medical place monitoring video frame to obtain a new medical place monitoring video frame with a corresponding number of frames for replacing the medical place monitoring video frame;

secondly, for each second target monitoring video, clustering the medical site monitoring video frames and/or the new medical site monitoring video frames (determined according to the actual video frames included in the second target monitoring video) included in the second target monitoring video based on whether the corresponding user objects are the same or not to obtain at least one corresponding video frame clustering set, and respectively clustering each video frame set into a monitoring video to be identified corresponding to one user object.

For example, in some preferred embodiments, the clustering, for each second target surveillance video, the medical-site surveillance video frames and/or the new medical-site surveillance video frames included in the second target surveillance video based on whether the corresponding user object is the same, to obtain at least one corresponding video-frame cluster set, and respectively clustering each video-frame cluster set as a surveillance video to be identified corresponding to one user object may include the following steps:

firstly, clustering the medical site monitoring video frames and/or the new medical site monitoring video frames included in each second target monitoring video based on whether the corresponding user objects are the same or not to obtain at least one corresponding video frame clustering set;

secondly, determining the monitored video to be identified corresponding to each user object according to each two video frame cluster sets in the at least one video frame cluster set and based on whether the user objects corresponding to the two video frame cluster sets are the same or not.

For example, in some preferred embodiments, the step of determining, for each two video frame cluster sets in the at least one video frame cluster set, the to-be-identified surveillance video corresponding to each user object based on whether the user objects corresponding to the two video frame cluster sets are the same may include the following steps:

firstly, determining whether user objects respectively corresponding to two video frame cluster sets are the same or not aiming at every two video frame cluster sets in the at least one video frame cluster set;

secondly, aiming at every two video frame cluster sets in the at least one video frame cluster set, if the user objects corresponding to the two video frame cluster sets are different, determining the two video frame cluster sets as two corresponding monitored videos to be identified respectively;

then, for every two video frame cluster sets in the at least one video frame cluster set, if the user objects corresponding to the two video frame cluster sets are the same, the two video frame cluster sets are merged into a corresponding monitoring video to be identified.

For example, in some preferred embodiments, the step of performing, for each of the at least one to-be-identified monitoring video, user behavior guidance processing on the user object in the to-be-identified monitoring video based on the medical site monitoring video frame included in the to-be-identified monitoring video, that is, step S300, may include the following steps:

firstly, aiming at each monitored video to be identified in at least one monitored video to be identified, performing action identification processing on the monitored video to be identified to obtain an action identification result corresponding to the monitored video to be identified, and determining whether user behavior guidance processing needs to be performed on a user object in the monitored video to be identified or not based on the action identification result;

secondly, for each to-be-identified surveillance video in the at least one to-be-identified surveillance video, if it is determined that user behavior guidance processing needs to be performed on the user object in the to-be-identified surveillance video, user behavior guidance processing (for example, prompting a corresponding site manager or medical care personnel to guide the user object) is performed on the user object in the to-be-identified surveillance video.

For example, in some preferred embodiments, the step of performing, for each to-be-identified surveillance video in the at least one to-be-identified surveillance video, action recognition processing on the to-be-identified surveillance video to obtain an action recognition result corresponding to the to-be-identified surveillance video, and determining whether user behavior guidance processing needs to be performed on a user object in the to-be-identified surveillance video based on the action recognition result may include the following steps:

firstly, aiming at each monitored video to be identified in at least one monitored video to be identified, performing action identification processing on the monitored video to be identified to obtain an action identification result corresponding to the monitored video to be identified, and determining whether a user object corresponding to the monitored video to be identified has repeated same actions (such as walking back and forth in an area) or not based on the action identification result;

secondly, for each monitored video to be identified in the at least one monitored video to be identified, if the user object corresponding to the monitored video to be identified has the same action repeated for multiple times, it is determined that user behavior guidance processing needs to be performed on the user object in the monitored video to be identified, and if the user object corresponding to the monitored video to be identified does not have the same action repeated for multiple times, it is determined that user behavior guidance processing does not need to be performed on the user object in the monitored video to be identified.

With reference to fig. 3, an embodiment of the present invention further provides a user behavior guidance system based on intelligent health monitoring, which can be applied to the medical monitoring server. Wherein, user's action bootstrap system based on wisdom health monitoring includes:

For example, in some preferred embodiments, the second video processing module is specifically configured to:

for each target monitoring video in the target monitoring videos, performing object identification processing on each frame of monitoring video frame of the medical place included in the target monitoring video to determine each user object corresponding to the target monitoring video; counting the number of user objects corresponding to the monitoring video frames of the medical places, which are included in the target monitoring video, aiming at each target monitoring video in the target monitoring videos to obtain the number of video frame objects corresponding to the target monitoring video; for each target surveillance video in the multiple target surveillance videos, determining a relative size relationship between the number of video frame objects corresponding to the target surveillance video and a preset object number threshold, and determining the target surveillance video as a first target surveillance video when the number of video frame objects is less than the object number threshold, or determining the target surveillance video as a second target surveillance video when the number of video frame objects is greater than or equal to the object number threshold; determining, for each of the first target surveillance videos, whether a user object in each of two frames of medical site surveillance video frames included in the first target surveillance video is the same when at least one of the first target surveillance videos exists, and when the user objects in every two frames of monitoring video frames of the medical place included in the first target monitoring video are the same, determining the first target monitoring video as the monitoring video to be identified, and screening out each medical site surveillance video frame having a user object when the user objects in at least two medical site surveillance video frames included in the first target surveillance video are different, classifying the monitoring video frames of the medical place based on whether the user objects are the same or not to obtain at least one corresponding video frame set, and respectively taking each video frame set as a monitoring video to be identified corresponding to a user object.

For example, in some preferred embodiments, the user behavior guidance module is specifically configured to:

performing action recognition processing on the surveillance video to be recognized aiming at each surveillance video to be recognized in the at least one surveillance video to be recognized to obtain an action recognition result corresponding to the surveillance video to be recognized, and determining whether user behavior guidance processing needs to be performed on a user object in the surveillance video to be recognized or not based on the action recognition result; and aiming at each monitored video to be identified in the at least one monitored video to be identified, if the user behavior guidance processing needs to be carried out on the user object in the monitored video to be identified, carrying out user behavior guidance on the user object in the monitored video to be identified.

In summary, the present invention provides a method and a system for guiding user behavior based on intelligent health monitoring, after the obtained multiple medical site monitoring videos collected by the multiple medical monitoring terminal devices are processed to obtain corresponding multiple target monitoring videos, each target monitoring video can be firstly targeted, performing user analysis processing on the target monitoring video to obtain a user analysis result corresponding to the target monitoring video, and processing the target monitoring video based on the user analysis result to obtain at least one to-be-identified monitoring video corresponding to the target monitoring video, so that the user behavior guiding processing can be carried out on the user object corresponding to each monitoring video to be identified based on the monitoring video to be identified, namely, the user behavior guidance processing is actively carried out on the user object, so that the problem of poor effect on user behavior guidance in the prior art is solved.

The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention, and various modifications and changes may be made by those skilled in the art. Any modification, equivalent replacement, or improvement made within the spirit and principle of the present invention should be included in the protection scope of the present invention.

Claims

1. A user behavior guiding method based on intelligent health monitoring is characterized by being applied to a medical monitoring server, the medical monitoring server is in communication connection with a plurality of medical monitoring terminal devices, and the user behavior guiding method based on intelligent health monitoring comprises the following steps:

2. The method as claimed in claim 1, wherein the step of performing user parsing on each target surveillance video of the plurality of target surveillance videos to obtain a user parsing result corresponding to the target surveillance video, and performing processing on the target surveillance video based on the user parsing result to obtain at least one surveillance video to be identified corresponding to the target surveillance video includes:

3. The method as claimed in claim 2, wherein the step of performing user analysis processing on each target surveillance video of the target surveillance videos to obtain a user analysis result corresponding to the target surveillance video, and performing processing on the target surveillance video based on the user analysis result to obtain at least one surveillance video to be identified corresponding to the target surveillance video further comprises:

4. The method as claimed in claim 3, wherein the step of clustering the medical-field surveillance video frames and/or the new medical-field surveillance video frames included in the second target surveillance video to obtain at least one corresponding video-frame cluster set based on whether the corresponding user object is the same for each second target surveillance video, and respectively clustering each video-frame cluster set as a surveillance video to be identified corresponding to a user object comprises:

5. The method as claimed in claim 4, wherein the step of determining the monitored video to be identified corresponding to each user object based on whether the user objects corresponding to the two video frame cluster sets are the same for each two video frame cluster sets in the at least one video frame cluster set comprises:

6. The method for guiding user behavior based on intelligent health monitoring as claimed in any one of claims 1 to 5, wherein the step of performing, for each monitored video to be identified in the at least one monitored video to be identified, user behavior guidance processing on the user object in the monitored video to be identified based on the monitoring video frame of the medical place included in the monitored video to be identified includes:

7. The method as claimed in claim 6, wherein the step of performing motion recognition processing on the surveillance video to be recognized to obtain a motion recognition result corresponding to the surveillance video to be recognized, and determining whether the user behavior guidance processing needs to be performed on the user object in the surveillance video to be recognized based on the motion recognition result comprises:

8. The utility model provides a user's action bootstrap system based on wisdom health monitoring which characterized in that is applied to medical monitoring server, medical monitoring server communication connection has a plurality of medical monitoring terminal equipment, user's action bootstrap system based on wisdom health monitoring includes:

9. The intelligent health monitoring-based user behavior guidance system of claim 8, wherein the second video processing module is specifically configured to:

10. The intelligent health monitoring-based user behavior guidance system of claim 8, wherein the user behavior guidance module is specifically configured to: